Pivotal role of DPYSL2A in KLF4-mediated monocytic differentiation of acute myeloid leukemia cells

Although the biological importance of Krüppel-like factor 4 (KLF4) transcription factor in the terminal differentiation of hematopoietic cells to the monocytes has been well established, the underlying mechanisms remain elusive. To clarify the molecular basis of KLF4-mediated monocytic differentiation, we performed detailed genetic studies in acute myeloid leukemia (AML) cells. Here, we report that dihydropyrimidinase like 2 (DPYSL2), also known as CRMP2, is a novel key differentiation mediator downstream of KLF4 in AML cells. Interestingly, we discovered that KLF4-mediated monocytic differentiation is selectively dependent on one specific isoform, DPYSL2A, but not on other DPYSL family genes. Terminal differentiation to the monocytes and proliferation arrest in AML cells induced by genetic or pharmacological upregulation of KLF4 were significantly reversed by short hairpin RNA (shRNA)-mediated selective depletion of DPYSL2A. Chromatin immunoprecipitation assay revealed that KLF4 associates with the proximal gene promoter of DPYSL2A and directly transactivates its expression. Together with the unique expression patterns of KLF4 and DPYSL2 limited to the differentiated monocytes in the hematopoietic system both in human and mouse, the identified KLF4-DPYSL2 axis in leukemia cells may serve as a potential therapeutic target for the development of novel differentiation therapies for patients with AML.


Scientific Reports
| (2020) 10:20245 | https://doi.org/10.1038/s41598-020-76951-0 www.nature.com/scientificreports/ a key role in the development of Alzheimer's disease 9,11,12 . Alternatively spliced transcript variants encoding multiple isoforms have been observed for this gene, but little is known about their roles in hematopoietic cells or the development and maintenance of leukemia.
Here, we show that KLF4-mediated terminal differentiation of AML cells into the monocytes is mediated by a specific isoform of the DPYSL2 gene, DPYSL2A, thus providing a basis for the development of novel differentiation therapies for AML treatment in the future.

Results
KLF4 directly transactivates the expression of DPYSL2A in AML cells. As we and others have previously reported, increased levels of KLF4 induce terminal differentiation of AML cells into the monocytic lineage 3,4 . We first confirmed that higher levels of KLF4 expression are restricted to the monocytes among all hematopoietic cell lineages in both humans and mice (Fig. 1A, Supplementary Figs. S1B). Consistent with a previous report 4 , the mRNA expression levels of KLF4 were significantly higher in differentiated monocytes than that in AML cells, suggesting that induction of the expression levels of KLF4 could be a therapeutic strategy for patients with AML (Fig. 1A). Indeed, exogenous expression of KLF4 significantly induced G 1 phase cell cycle arrest and suppressed the growth of leukemia cells in two different AML cell lines, THP-1 and KO52 (Fig. 1B, AML t(11q23)/MLL, n = 43; AML complex, n = 48) or normal hematopoietic cells of various lineages (HSC hematopoietic stem and progenitor cells, n = 6; MPP multipotent progenitors, n = 2; CMP common myeloid progenitors, n = 3; GMP granulocyte-monocyte progenitors, n = 7; MEP megakaryocyte-erythrocyte progenitors, n = 4; early_PM early promyelocytes, n = 3; late_PM late promyelocytes, n = 3; MY myelocytes, n = 2; MM metamyelocytes, n = 3; BC band cells, n = 4; PMN polymorphonuclear cells, n = 3; Mono Monocytes, n = 4) (GSE42519 and GSE13159). Data were retrieved from the BloodSpot database 30 . The database is freely available at www.blood spot.eu. (B) Cell proliferation curves of THP-1 cells transduced with a lentivirus encoding KLF4 or control cassette. Cells were cultured in the presence of 3 μM doxycycline (n = 3). (C) Representative microscopic images of THP-1 cells, as in (B). Cells were treated with 3 μM doxycycline for the indicated time periods, harvested, and cytospun onto glass slides. Diff-Quik staining (modified Giemsa staining) was performed on each of the slides (original magnification: ×20, scale bar 50 μm). Data are presented as mean ± SEM. ***P < 0.001, by two-tailed Student's t test. ). KLF4 expression also led to the terminal differentiation of these AML cells to the monocytes (Fig. 1C, Supplementary Fig. S2B).
To identify essential differentiation mediators downstream of KLF4 in AML cells, we first analyzed the previously reported gene expression microarray data sets. To begin with, we compared the expression levels of each of the genes in mouse myeloid progenitor Tot2 cells with and without exogenous Klf4 expression (GSE38810) and extracted the top 1000 upregulated genes associated with exogenous Klf4 expression. Interestingly, continuous stimulation of the mitogen-activated protein kinase (MAPK) signaling cascade has been shown to activate hematopoietic stem and progenitor cells (HSPCs) with increased numbers of differentiated progenitors 13 . Because chronic activation of MAPK signaling has also been shown to induce KLF4 expression 2,3 , we next analyzed the expression levels of each of the genes in mouse HSPCs with somatic Nras mutation (GSE45194). We compared the expression levels of each of the genes from wild-type and Nras mutation + mice and extracted the top 1000 upregulated genes associated with the Nras mutation. In addition, we analyzed the expression levels of each of the genes in human hematopoietic cells from the bone marrow of 154 primary de novo AML patients (GSE22845). We divided the patients into two groups according to their KLF4 expression levels and extracted the top 1,000 upregulated genes that are associated with KLF4 high-expressing AML patients. A Venn diagram was used to identify the overlapping genes in these three data sets, and 26 genes that are the potential downstream targets of KLF4 transcription factor were identified ( Fig. 2A; Supplementary Table S1). We then performed qRT-PCR analysis and examined the changes in the expression levels of each of these 26 genes upon exogenous KLF4 expression in THP-1 AML cells. Results showed that 16 out of the 26 genes were significantly upregulated upon forced KLF4 expression. Among the upregulated genes, the expression levels of DPYSL2 were substantially elevated over 2000-fold relative to the baseline (Fig. 2B). Since previous reports have suggested its vital role in neuronal differentiation and polarity as well as axonal growth and guidance, we speculated the possible involvement of DPYSL2 in the regulation of KLF4-mediated monocytic differentiation in AML cells 10,14 . Indeed, the expression levels of DPYSL2 were significantly higher in differentiated monocytes than those in other lineage cells or AML cells (Fig. 2E, Supplementary Fig. S4A,B).
The DPYSL family of proteins is encoded by seven distinct genes, DPYSL1, DPYSL2A (DPYSL2 isoform 1), DPYSL2B (DPYSL2 isoform 2), DPYSL2C (DPYSL2 isoform 3), DPYSL3, DPYSL4, and DPYSL5. DPYSL2B and DPYSL2C differ in the 5′ untranslated region (5′UTR) compared to DPYSL2A and lack a portion of the 5′-coding region, and thus have a shorter N-terminus. Interestingly, qRT-PCR assay in THP-1 and KO52 cells demonstrated that forced expression of KLF4 selectively induced the expression of DPYSL2A isoform among the other DPYSL family members (Fig. 2C, Supplementary Fig. S5A). As shown in Fig. 2D and Supplementary Fig. S5B, the protein expression levels of DPYSL2A were also specifically and significantly increased upon exogenous KLF4 expression in THP-1 and KO52 cells, without evident induction of the other isoforms of this gene, DPYSL2B and DPYSL2C. Of note, we identified that the mRNA expression levels of KLF4 were positively correlated to those of DPYSL2A, but not to DPYSL2B, in 14 different AML cell lines ( Supplementary Fig. S5C). In addition, by analyzing five independently published gene expression data sets (GSE12417, GSE15434, GSE21261, GSE22845 and GSE37642), we found that the mRNA expression levels of KLF4 and DPYSL2 are positively correlated in primary AML cells (Fig. 3A, Supplementary Fig. S6). Analysis of the proximal DPYSL2A promoter region [− 1500 bp to + 250 bp of transcription start site (TSS)] revealed multiple KLF4 core consensus binding sequences 15,16 Supplementary Fig. S7). ChIP assay using anti-KLF4 antibody confirmed the direct association between KLF4 and these sites, indicating the direct transactivation of DPYSL2A by KLF4 protein (Fig. 3C).
Next, we pharmacologically modulated the expression levels of KLF4 within the physiological range using phorbol 12-myristate 13-acetate (PMA) in THP-1 cells. PMA has been shown to induce rapid and sustained activation of Ras-Raf-MEK-ERK pathway, eventually inducing terminal differentiation of AML cells to monocytes through upregulation of KLF4 expression 4 . Consistent with previous reports, PMA treatment rapidly induced KLF4 and DPYSL2A expression in THP-1 cells under our experimental conditions within 24 h (Fig. 4D). Furthermore, continuous exposure to PMA significantly suppressed the growth of THP-1 cells. In contrast, the proliferation of THP-1 cells depleted of KLF4 or DPYSL2A expression was not affected by PMA treatment (Fig. 4E). These results collectively indicate that DPYSL2A is a key mediator of terminal differentiation to monocytes downstream of KLF4.

DPYSL2A, but not DPYSL2B, mediates monocytic differentiation in AML cells.
To further examine the role of DPYSL2A in AML cells, we prepared THP-1 cells expressing either DPYSL2A or DPYSL2B (Fig. 5A,B, Supplementary Fig. S9). As shown in Fig. 5C,D, exogenous expression of DPYSL2A resulted in monocytic differentiation of THP-1 cells and suppressed cell growth. In contrast, exogenous expression of DPYSL2B, another isoform of DPYSL2 that lacks 118 amino acids in its N terminus, neither induced monocytic differentiation of THP-1 cells nor suppressed cell growth (Fig. 5C,D)

Discussion
In adults, besides terminally differentiated monocytes 4 , KLF4 is predominantly expressed in the epithelial cells of the gastrointestinal tract 17 , skin 18 , vascular endothelial cells 19 and thymus 20 . In addition to its documented effect on cell self-renewal and proliferation 21 , KLF4 has been shown to regulate tissue differentiation in these organs 22 .
For example, a ~ 90% decrease in the number of differentiated goblet cells was observed in the gastrointestinal tract in Klf4 knockout mice, without any abnormal changes in the frequency of other cell populations 23 . In addition, in newborn Klf4 −/− mice, terminal differentiation of the epidermis was disturbed, and mice died shortly after birth due to their inability to maintain a skin barrier and rapid loss of body fluids 18 . These results suggest the crucial role of KLF4 in the differentiation of certain cells, such as colonic epithelial cells and epidermal cells.
In this study, we demonstrated the essential role of the transcription factor KLF4 in the terminal differentiation of hematopoietic cells toward the monocytic lineage. The most striking result we present here is that the ability of KLF4 to promote terminal differentiation to the monocytes is highly dependent on the single specific isoform of DPYSL2. Since DPYSL2B, not DPYSL2A is the more abundant isoform of DPYSL2 in the tissues studied so far, previous studies have investigated the function of the DPYSL2B isoform [7][8][9][10] . Indeed, the biological role of DPYSL2B is well established in neuronal cells, where it regulates axonal growth and differentiation by directly interacting with cytoskeletal proteins, such as tubulin and dynein 7,8 . DPYSL2A and DPYSL2B transcripts use distinct promoters and differ in their N-terminal amino acid sequences. The DPYSL2B transcript encodes a protein of 572 amino acids, whereas the DPYSL2A transcript encodes a longer protein of 677 amino acids. The two DPYSL2 isoforms have an identical C-terminal 559 amino acids sequence and differ only in their N-terminals. Considering that exogenous expression of only DPYSL2A induces monocytic differentiation in AML cells, the unique N-terminal sequence of 118 amino acids in DPYSL2A might play important roles in mediating this process in hematopoietic cells, but not other cells, such as neuronal cells, where DPYSL2B is dominant and mediates cell differentiation 24 .
From a clinical perspective, our findings provide significant insights for translational researchers to develop novel therapeutic strategies for the treatment of patients with AML. Since recent discoveries and the emergence of new therapeutic agents have bypassed AML cases with poor prognosis, development of alternative therapeutic strategies based on novel drug mechanisms is imperative for this subset of patients. The use of all-trans retinoic acid and arsenic trioxide in acute promyelocytic leukemia has proven that differentiation therapy may significantly improve the survival of patients with AML; however, its success has not been translated to other groups of AML 25 . Therefore, the identification of new therapeutic agents that may induce the differentiation of AML blasts represents an attractive new target, such as the KLF4-DPYSL2 axis identified in this study. Currently, there are no clinically applicable agents available to stimulate this axis. The only experimentally validated KLF4-inducing compound with differentiation potential toward monocytes in AML cells is PMA, which is unsuitable for clinical use due to its serious systemic side effects 26 . This immediate demand for KLF4-DPYSL2A axis-activating compounds should be addressed in future studies, which may help treat otherwise incurable subsets of AML patients.
Taken together, we identified the crucial role of the KLF4-DPYSL2A axis in the terminal differentiation of AML cells to the monocytes in this study. Further studies are required for the better understanding of its precise underlying molecular mechanism. Moreover, developing tools to pharmacologically stimulate the KLF4-DPYSL2A axis in leukemia cells should be enthusiastically explored to establish a novel "differentiation therapy" for patients with AML.  30 . The database is freely available at www.blood spot.eu. Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, by two-tailed Student's t test.

Real-time quantitative polymerase chain reaction (qRT-PCR).
Total RNA was isolated using the RNeasy mini kit (Qiagen, USA) and reverse-transcribed with a reverse script kit (TOYOBO, Japan) to generate cDNA. qRT-PCR was performed on a 7500 Real-Time PCR System (Applied Biosystems, USA) according to the manufacturer's recommendations. The results were normalized to the expression levels of glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Relative expression levels were calculated using the 2 -ΔΔCt method 27 . The primers used for qRT-PCR are listed in Supplementary Table S3.
Chromatin immunoprecipitation-quantitative polymerase chain reaction (ChIP-qPCR). ChIP assay was performed using the SimpleChIP Plus Enzymatic Chromatin IP Kit (Cell Signaling Technology, USA) according to the manufacturer's protocol. In brief, cells were cross-linked in 1% formaldehyde in phosphate buffered saline for 10 min at room temperature. After glycine quenching, cell pellets were collected, lysed, and then subjected to sonication with the Q55 sonicator system (QSONICA, USA). The supernatant was diluted with the    Immunoblotting. Immunoblotting was performed as described previously 3  Cell cycle assay. THP-1 cells transduced with the lentivirus encoding KLF4 were treated with 3 μM doxycycline to induce gene expression. Twenty-four hours after the treatment, cells were fixed in 100% ethanol followed by RNA digestion in PBS containing 100 μg/mL of RNase A. Samples were then stained with 50 μg/mL propidium iodide and immediately subjected to flow cytometric analysis.
Gene expression microarray analysis. THP-1 cells transduced with the lentivirus encoding KLF4 or DPYSL2A were treated with 3 μM doxycycline for 48 h to induce gene expression. After the indicated incubation time, total RNA was prepared, and its quality was assessed using an Agilent 2100 Bioanalyzer (Agilent Technologies, USA). Using 250 ng of total RNA, biotinylated cRNA was prepared according to the standard Affymetrix protocol with the GeneChip 3′ IVT PLUS Reagent Kit. Following fragmentation, 15 µg of cRNA was hybridized for 16 h at 45 °C on the GeneChip Hybridization Oven 645. GeneChips were washed and stained in the GeneChip Fluidics Station 450 using GeneChip Hybridization, Wash, and Stain Kit. GeneChips were scanned using the GeneChip Scanner 3000 7G. The data were analyzed with Microarray Suite version 5.0 (MAS 5.0) using Affymetrix default analysis settings and global scaling as a normalization method. The trimmed mean target intensity of each array was arbitrarily set to 100. Our microarray data have been deposited in NCBI's Gene Expression Omnibus (GEO Series accession number GSE101751). Gene set enrichment analysis (GSEA) was utilized to analyze the microarray data obtained in the present study 29 . Statistical analysis. Differences between control and experimental groups were assessed by a two-tailed unpaired Student's t test and declared statistically significant if the p value was less than 0.05. The equality of variances in two populations was calculated using the F test. The results are represented as the mean ± standard error of mean (SEM) of values obtained from three independent experiments. In transplantation experiments, animals were randomly allocated to the experimental groups and the treatments were given by blinding. Survival rates between the indicated groups were compared using the log-rank test.