CEBPG promotes acute myeloid leukemia progression by enhancing EIF4EBP1

Background Acute myeloid leukemia (AML) is a myeloid neoplasm accounts for 7.6% of hematopoietic malignancies. AML is a complex disease, and understanding its pathophysiology is contributing to the improvement in the treatment and prognosis of AML. In this study, we assessed the expression profile and molecular functions of CCAAT enhancer binding protein gamma (CEBPG), a gene implicated in myeloid differentiation and AML progression. Methods shRNA mediated gene interference was used to down-regulate the expression of CEBPG in AML cell lines, and knockdown efficiency was detected by RT-qPCR and western blotting. The effect of knockdown on the growth of AML cell lines was evaluated by CCK-8. Western blotting was used to detect PARP cleavage, and flow cytometry were used to determine the effect of knockdown on apoptosis of AML cells. Genes and pathways affected by knockdown of CEBPG were identified by gene expression analysis using RNA-seq. One of the genes affected by knockdown of CEBPG was Eukaryotic translation initiation factor 4E binding protein 1 (EIF4EBP1), a known repressor of translation. Knockdown of EIF4EBP1 was used to assess its potential role in AML progression downstream of CEBPG. Results We explored the ChIP-Seq data of AML cell lines and non-AML hematopoietic cells, and found CEBPG was activated through its distal enhancer in AML cell lines. Using the public transcriptomic dataset, the Cancer Cell Line Encyclopedia (CCLE) and western blotting, we also found CEBPG was overexpressed in AML. Moreover, we observed that CEBPG promotes AML cell proliferation by activating EIF4EBP1, thus contributing to the progression of AML. These findings indicate that CEBPG could act as a potential therapeutic target for AML patients. Conclusion In summary, we systematically explored the molecular characteristics of CEBPG in AML and identified CEBPG as a potential therapeutic target for AML patients. Our findings provide novel insights into the pathophysiology of AML and indicate a key role for CEBPG in promoting AML progression. Supplementary Information The online version contains supplementary material available at 10.1186/s12935-021-02305-z.


Introduction
Acute myeloid leukemia (AML) is a myeloid neoplasm that accounts for 7.6% of hematopoietic malignancies. It is caused by the oncogenic transformation of hematopoietic progenitors in the bone marrow (BM), which results in the destruction of blood tissue. AML is reported to have a long-term survival of less than 20% [1][2][3]. Every year there are about 18,000 new cases AML in Europe
Among CEBPs, CEBPA, CEBPE and CEBPZ have been reported to function in AML development [9,19,20], however the role of CCAAT enhancer binding protein gamma (CEBPG) in AML is unclear. CEBPG is a member of leucine-zipper transcription factor family that plays a role in many biological processes [21][22][23][24]. Knockdown of CEBPG suppressed tumor growth [25]. CEBPG is suggested as a biomarker for lung cancer risk [26]. It is also involved in the differentiation arrest in AML [27,28]. Although the roles of CEBPG in several types of cancer have been revealed, its expression profile and molecular functions in AML remain unresolved. Therefore, in this study we assess the role of CEBPG in AML progression.
In the present study, shRNA mediated gene interference was used to down-regulate the expression of CEBPG in AML cell lines, and the knockdown efficiency was detected by RT-qPCR and western blotting. The effect of CEBPG knockdown on the growth of AML cell lines was evaluated by Cell Counting Kit-8 (CCK-8) assays. Western blotting was used to detect poly(ADPribose) polymerase (PARP) cleavage, and flow cytometry was used to determine the effect of CEBPG knockdown on apoptosis of AML cells. Genes and pathways affected by knockdown of CEBPG were identified by gene expression analysis using RNA-seq.
One of the genes affected by knockdown of CEBPG was Eukaryotic translation initiation factor 4E binding protein 1 (EIF4EBP1). EIF4EBP1 is a translation repressor protein [29] that plays a role in multiple types of cancer, including lung, breast, and liver cancer [30][31][32][33]. For example, EIF4EBP1 is reported to be significantly overexpressed in hepatocellular carcinoma (HCC) tissues and is related to poor survival of patients with HCC [33]. However, the biological effect and underlying mechanism of EIF4EBP1 in AML has not been explored. Therefore, knockdown of EIF4EBP1 was used to assess its potential role in AML progression downstream of CEBPG.
In the present study, we explored the ChIP-Seq data of AML cell lines and non-AML hematopoietic cells and found CEBPG was activated through its distal enhancer in AML cell lines. Using the public transcriptomic dataset, the Cancer Cell Line Encyclopedia (CCLE) and western blotting, we also found that CEBPG was overexpressed in AML. Moreover, CEBPG promotes AML cell proliferation by activating EIF4EBP1, thus contributing to the progression of AML. These findings indicate that CEBPG could act as a potential therapeutic target for AML patients.

Cell lines and culture
Human AML cell lines, including NB4,THP-1, MV4-11, and K562 which was from blastic crisis of chronic myelogenous leukemia were obtained from the cell bank of the American type culture collection and cultured in RPMI medium (Termo Fisher Scientifc) containing 10% fetal bovine serum (Biological Industries, CT, USA), and 1% penicillin-streptomycin (Beyotime Biotechnology, Shanghai, China) at 37 °C in a humidified incubator with an atmosphere of 5% CO2 and tested routinely for mycoplasma.

Cell viability assay
Leukemia cells were seeded in 96-well plates at a density of 1 × 10 3 cells per well. The cell viability was determined by Cell Counting kit-8 (CCK8) assay (Dojindo Molecular Technologies, Tokyo, Japan) according to the manufacturer's instructions. Cell proliferation was calculated as a percentage of that in cells in control medium. Each concentration was tested in triplicate and repeated in at least three independent experiments. The calculation was performed by Graph Prism software 7.0 (GraphPad Software Inc., San Diego, CA, USA).

RNA preparation and real-time PCR expression analysis
Total RNA was extracted from cell pellets using TRIzol ® reagent (Invitrogen, CA, USA), according to the manufacturer's protocol. For cDNA synthesis, 1 µg of total RNA was converted to cDNA using a High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, CA, USA). Quantitative real-time PCR analysis was carried out using LightCycler ® 480 SYBR Green I Master mix (cat. No. 04707516001; Roche, Penzberg, Germany) with a LightCycler 480 Real Time System (Roche), according to the manufacturer's protocol. mRNA expression levels were calculated using the Ct method with glyceraldehyde 3-phosphate dehydrogenase (GAPDH) expression as an internal reference. Primer sequences are listed in Table 2.

Western blotting analysis
Western blotting analysis was conducted using the following primary antibodies: CEBPG (cat.

RNA-seq and data processing
RNA-seq was carried out according to the protocols suggested by Novogene, Beijing, China. First, total RNA was reverse transcribed to cDNA for library construction, and the cDNA library was then sequenced. The raw reads were filtered and clean reads were mapped according to HISAT. The gene expression level (as fragments per kilobase of exon model per million reads mapped) was then calculated. Differentially expressed genes (P < 0.05 and fold-change > 2 or fold-change < 0.5) were identified using DESeq2 analysis. For enrichment analysis, differentially expressed genes were analyzed using the DAVID Bioinformatics Resources v6.8 online server (https:// david. ncifc rf. gov).  [34], with alignment parameters -p 4 -q -x. Peaks were identified using MACS2 (v2.0.9) [35], with parameters -g hs -n test -B -q 0.01. The bedgraph files generated by MACS2 were converted to bigwig files using the UCSC bedGraphToBigWig tool, and then bigwig files were visualized by Integrative Genomics Viewer (IGV) [36].

Statistical analysis
The association between EIF4EBP1 expression and overall survival of AML patients were assessed using the Kaplan-Meier analysis. Comparison between two groups was carried out using the Student's t-test or the Mann-Whitney u test. Statistical analysis was carried out by GraphPad Prism 7.0 (GraphPad Software, Inc., La Jolla, CA, USA). Statistically significant P values are indicated as *P < 0.05, ** P < 0.01, ***P < 0.001, and ****P < 0.0001.

CEBPG is activated through its distal enhancer and is overexpressed in AML cell lines
By interrogating ChIP-Seq data of AML cell lines (Fig, 1a, tracks 1-6, K562 cell line also included) and non-AML hematopoietic cells (Fig. 1a, tracks 7-10), we found that the enhancer region of CEBPG in AML cell lines showed coincident H3K27ac signals that were not present in non-AML hematopoietic cells, suggesting a potential role in transcription regulation. Then, we assessed the expression pattern of CEBPG between AML patients and healthy controls in a public transcriptomic dataset (GSE114868) [37], and found that CEBPG was more highly expressed in AML samples (Fig. 1b) relative to that in healthy control samples (the differentially expressed genes between AML and control samples in dataset GSE114868 are listed in Additional file 1: Table S1). Moreover, the Cancer Cell Line Encyclopedia (CCLE; https:// porta ls. broad insti tute. org/ ccle) which includes CEBPG mRNA expression profiles for multiple cancer cell lines, showed that CEBPG was highly expressed in hematologic malignancies including AML (Fig. 1c). We also assessed the levels of CEBPG in AML and non-AML cell lines using western blotting, and found higher levels of CEBPG in AML cell lines than in non-AML cell lines ( Fig. 1d and e). Collectively, these data suggested that CEBPG is activated through its distal enhancer and overexpressed in AML.

CEBPG is oncogenic and promotes AML cell proliferation
To address the biological significance of CEBPG, we selected three AML cell lines with high CEBPG protein levels shown in Fig. 1d (THP-1, NB4 and MV4-11) and performed shRNA-mediated knockdown of CEBPG using three independent shRNAs (Table 1). Knockdown efficiency of CEBPG was evaluated using western blotting and qPCR (Fig. 2a, b, e and f ). Notably, knockdown of CEBPG significantly inhibited the proliferation rates of all 3 AML cell lines (Fig. 2c, d, g and h). We also assessed the level of the apoptotic protein PARP using western blotting and found that PARP levels were increased in both MV4-11 and NB4 cell lines upon knockdown of CEBPG (Fig. 2i). Knockdown of CEBPG also increased the apoptotic rates of MV4-11 and NB4 cell lines (Fig. 2j  and k). Altogether, these data suggested that CEBPG is oncogenic and contributes to the proliferation of AML cells.

CEBPG activates EIF4EBP1 in AML cell lines
To reveal potential targets responsible for CEBPG-promoted AML cell proliferation, RNA-seq analyses were performed on NB4 and MV4-11 cell lines comparing shRNA control cells with CEBPG knockdown cells. A total of 1196 and 2207 differently expressed genes (DEGs) were identified upon CEBPG knockdown, in NB4 and MV4-11 cell lines respectively (Log2 |fold change|> 1, P < 0.05, Fig. 3a and b). EIF4EBP1 was included in the top 10 downregulated genes upon CEBPG knockdown in both NB4 and MV4-11 cell lines ( Fig. 3c and d). Next, we conducted a functional enrichment analysis of all DEGs using the KEGG Pathway Database. The results showed a significant enrichment for EGFR tyrosine kinase inhibitor resistance signaling (ranking 4 th ), which involves EIF4EBP1 (Fig. 3e). Therefore, EIF4EBP1 was selected for in-depth investigation. To further determine the regulation of CEBPG on EGFR tyrosine kinase inhibitor resistance signaling and EIF4EBP1, a total of 8 genes (EIF4EBP1, PLCG1, EIF4E, AXL, PIK3R2, MET, PDGFB and SRC) from the EGFR tyrosine kinase inhibitor resistance signaling pathway was selected for qRT-PCR validation. In accordance with the RNA-Seq results, the mRNA levels of 6 of these genes, including EIF4EBP1, were downregulated while 2 genes were upregulated in NB4 cells in response to CEBPG silencing ( Fig. 3f and g). Additionally, ChIP-Seq data of AML cell lines and K562 cell line showed that the promoter region of EIF4EBP1 had coincident H3K27ac signals (Fig. 3h, tracks 1-6), while the ChIP-Seq data from K562 cells further indicated that EIF4EBP1 was bound by CEBPG at its TSSproximal regions (Fig. 3h, track 7), suggesting a potential role for CEBPG in the transcriptional regulation of EIF-4EBP1. Therefore, we next investigated the role of EIF-4EBP1 in NB4 and K562 cells.

EIF4EBP1 knockdown interferes with AML cell proliferation and increases apoptosis
To evaluate the biological significance of EIF4EBP1, we selected 2 cell lines (NB4 and K562) and knocked down EIF4EBP1 in both cell lines using three independent shRNAs (Table 1). Knockdown efficiency of EIF4EBP1 was evaluated using western blotting and qPCR (Fig. 4a,  b, h and i). Notably, knockdown of EIF4EBP1 significantly inhibited the proliferation rates of both cell lines (Fig. 4c, d, j and k). We also assessed the expression level of the apoptotic protein PARP using western blotting and found that PARP levels increased in both NB4 and K562 cell lines upon knockdown of EIF4EBP1 (Fig. 4e  and l). Knockdown of EIF4EBP1 also increased the apoptotic rates of NB4 and K562 cell lines (Fig. 2f, g, 4m and n). Collectively, these data suggested that EIF4EBP1 is required to sustain proliferation and survival of AML cells.

Identification of EIF4EBP1 as an unfavorable prognostic factor for AML patients
We assessed the expression pattern of EIF4EBP1 between AML patients and healthy controls in two public transcriptomic datasets (GSE114868 and GSE142700) [37]. The results showed that EIF4EBP1 was significantly overexpressed in AML samples in both datasets ( Fig. 5a and  b). To further explore the prognostic value of EIF4EBP1, we used the online tool http:// gepia. cancer-pku. cn/ and the result showed that the overall survival of AML patients with higher EIF4EBP1 expression was significantly poorer than those with lower EIF4EBP1 expression (Fig. 5c). These results suggested that EIF4EBP1 represents a negative prognostic factor for AML patients.

Discussion
AML is an aggressive malignancy with poor prognosis [8]. It is a complex disease and a detailed understanding of its pathophysiology is required to improve the treatment and prognosis of AML [5][6][7][8].
CCAAT enhancer binding proteins (CEBPs) including CEBPA, CEBPB, CEBPD, CEBPE, CEBPG and CEBPZ, are suggested as potential biomarkers for cancer prognosis [9][10][11][12][13][14]. Among CEBPs, CCAAT enhancer binding protein gamma (CEBPG), a member of leucine-zipper  transcription factor family, has been implicated in multiple cancers [25][26][27][28]. For example, it is reported that CEBPG significantly promotes the proliferation and migration of esophageal squamous cell carcinoma (ESCC) cells, and is thus suggested as a prognostic factor for patients with ESCC [21]. Although a role for CEBPG in myeloid differentiation has been demonstrated [27,28], if and how it contributes to the pathogenesis of AML is unclear. Here, we explored the function of CEBPG in AML and found that CEBPG is upregulated in AML and contributes to the proliferation of AML cells. We also demonstrated that CEBPG promotes AML cell proliferation by activating EIF4EBP1 in AML cell lines.
Eukaryotic translation initiation factor 4E binding protein 1 (EIF4EBP1) gene encodes a translation repressor protein [29]. This protein plays a role in multiple cancer types, including lung, breast, and liver cancer [30][31][32][33]. For example, EIF4EBP1 is reported to be significantly overexpressed in hepatocellular carcinoma (HCC) tissues and is related to poor survival of HCC patients [33]. However, the biological effect and underlying mechanism of EIF4EBP1 in AML has not been explored. In this study, we found the knockdown of EIF4EBP1 significantly inhibited proliferation and increases apoptosis in NB4 and K562 cells. Furthermore, in two public transcriptomic datasets (GSE114868 and GSE142700) [37], EIF4EBP1 was observed to be significantly overexpressed in AML samples. EIF4EBP1 was also identified as an unfavorable prognostic factor for AML patients using the online tool http:// gepia. cancer-pku. cn/. Taken together, these results suggested that EIF4EBP1 is involved in the pathogenesis of AML and represents a negative prognostic factor for AML patients.
In summary, we explored the function of CEBPG in AML and identified CEBPG as a potential therapeutic target for AML patients. Our findings provide novel insights into the pathophysiology of AML and elucidated a crucial role of CEBPG in promoting AML progression.
Additional file 1: Table S1. Differentially expressed genes between AML and control samples in dataset GSE114868. Fig. 5 a EIF4EBP1 was significantly overexpressed in AML samples in public transcriptomic dataset GSE114868. b EIF4EBP1 was significantly overexpressed in AML samples in public transcriptomic dataset GSE142700. c Base on the online tool http:// gepia. cancer-pku. cn/, the overall survival of AML patients with higher EIF4EBP1 expression was significantly poorer than those with lower EIF4EBP1 expression