Ubiquitous Expression of MAKORIN-2 in Normal and Malignant Hematopoietic Cells and Its Growth Promoting Activity

Makorin-2 (MKRN2) is a highly conserved protein and yet its functions are largely unknown. We investigated the expression levels of MKRN2 and RAF1 in normal and malignant hematopoietic cells, and leukemia cell lines. We also attempted to delineate the role of MKRN2 in umbilical cord blood CD34+ stem/progenitor cells and K562 cell line by over-expression and inhibition of MKRN2 through lentivirus transduction and shRNA nucleofection, respectively. Our results provided the first evidence on the ubiquitous expression of MKRN2 in normal hematopoietic cells, embryonic stem cell lines, primary leukemia and leukemic cell lines of myeloid, lymphoid, erythroid and megakaryocytic lineages. The expression levels of MKRN2 were generally higher in primary leukemia samples compared with those in age-matched normal BM cells. In all leukemia subtypes, there was no significant correlation between expression levels of MKRN2 and RAF1. sh-MKRN2-silenced CD34+ cells had a significantly lower proliferation capacity and decreased levels of the early stem/progenitor subpopulation (CFU-GEMM) compared with control cultures. Over-expression of MKRN2 in K562 cells increased cell proliferation. Our results indicated possible roles of MKRN2 in normal and malignant hematopoiesis.


Introduction
Makorin-2 (HSPC070; MKRN2) belongs to the MKRN gene family of which the ribonucleoproteins are characterized by a variety of zinc-finger motifs [1,2]. MKRN2 holds four C 3 H zinc fingers and a signature C 3 HC 4 RING zinc finger domain. MKRN2 is a highly conserved gene [1] yet its function remains largely unknown. Previous studies reported that mkrn2 in Xenopus laevis acted upstream of glucogen synthase kinase-3b in the phosphatidylinositol 3-kinase/Akt pathway. The third C 3 H zinc finger and the RING motif are required for the anti-neurogenesis activity [3,4]. MKRN2 was first identified in human CD34 + stem/ progenitor cells, as well as in some leukemic cell lines [2,5,6]. In chromosome 3p25, MKRN2 is located next to the proto-oncogene RAF1. Interestingly, they share a sequence of 105 bp in the 3' UTR in a reversed transcription orientation [2]. This antisense sequence-overlapping of MKRN2 with RAF1 suggested that these two genes may regulate each other and be involved in normal hematopoietic and leukemic development. In this study, we investigated the expression levels of MKRN2 and RAF1 in normal and malignant hematopoietic cells, and leukemic cell lines. We also attempted to explore the role of MKRN2 in umbilical cord blood CD34 + stem/progenitor cells and K562 cell line by overexpression and inhibition of MKRN2 through lentivirus transduc-tion and shRNA nucleofection, respectively. Our results demonstrated ubiquitous mRNA expression of MKRN2 and RAF1 in normal hematopoietic cells, embryonic stem cell lines, primary leukemia and leukemic cell lines. We also showed MKRN2 functions on promoting cell proliferation of primary CD34 + progenitor cells and K562 cells, indicating its possible involvement in normal and malignant hematopoiesis.

Ethics statement
Written informed consents were obtained for collection of all human samples. For minors/children enrolled in the study, written consents were obtained from their parents on their behalf. This study was approved by the Ethics Committee for Clinical Research of The Chinese University of Hong Kong. All necessary permits were obtained for the described study, which complied with all relevant regulations.

Patients and samples
Primary leukemic cells (over 70% blast cells) were obtained from the bone marrow of children (age #19 years) who were newly diagnosed with chronic myeloid leukemia (CML), acute lymphoid (ALL) or acute myeloid (AML) leukemia at the Prince of Wales Hospital, The Chinese University of Hong Kong. Age-matched normal subjects were siblings of patients who donated bone marrow for transplantation.
Peripheral blood samples were collected from normal adult volunteers. Mononuclear cells (MNC) were enriched by Ficoll-Hypaque density gradients (Amersham, Piscataway, NJ, USA). Human umbilical cord blood (CB) MNC and enriched CD34 + cells were obtained from full-term deliveries as described previously [7].

Over-expression of MKRN2 in CD34 + and K562 cells by lentivirus transduction
Full length makorin cDNA was subcloned into lentiviral vector (pLEF1aIG-MKRN2) (Fig A in File S1). The empty vector (pLEF1a-IG) was used as a control. The VSV-G pseudotyped lentivirus was produced by cotransfecting 293T cells with the transfer vector and three packaging vectors [10]. CD34 + or K562 cells were infected by the lentivirus at the multiplicity of infection (MOI) of 30. K562 cells were selected as the study model because K562 blasts are multipotential, hematopoietic malignant cells that could spontaneously differentiate into recognizable progenitors of the erythrocytic, granulocytic and monocytic lineages. In addition, we observed that MKRN2 and RAF1 were consistently and highly expressed in K562 and in primary myelocytic leukemia cells. Cells were first transduced for 16 hr, followed by 12 hr recovery in medium with 10% FBS and then transduced for a further 16 hr. Analysis and further manipulation was conducted 48 hr post transduction. The percentage of GFP expressing cells was monitored by flow cytometry using FACS Calibur flow cytometer and the CellQuest software (BD Biosciences), with 7-aminoactinomycin D (7-AAD) staining to gate out dead cells.

Silencing of MKRN2 by nucleofection using shRNA
A set of 29 mer shRNA constructs targeting MKRN2 (pGFP-V-RS-MKRN2, 4 unique sh cassettes in retroviral GFP vector) was introduced to inhibit the expression of MKRN2 in primary CD34 + cells and K562 cells. pGFP-V-RS (retroviral GFP vector) and pGFP-V-RS-NE (non-effective 29-mer sh GFP cassette retroviral GFP vector) were used as control experiments. All Hush constructs were purchased from OriGene Technologies (Rockville, MD, USA). Briefly, 200 ng of each of the shRNA plasmids were used for nucleofection. Enriched CB CD34 + cells (2610 4 /mL) and K562 cells (1610 5 /mL) were transfected using the Human CD34 + Cell Nucleofection Kit and K562 Nucleofection Kit (Amaxa Biosystems, Koeln, Germany), respectively. After nucleofection, cells were allowed to grow for 48 hr prior to measurement of readout parameters. The stable suppression of MKRN2 in K562 cells was maintained using Puromycin treatment (1 mg/mL; Invitrogen).

Cell viability
Transduced cells from each treatment were plated in duplicate wells (12-well plates, Corning) with the appropriate culture conditions (starting at 2610 4 cells/mL) and splitted at a ratio of 1:3 on day 6. Cells were counted daily by a hematocytometer under light-microscope, with trypan-blue staining (0.4%; Bio-Rad, Hercules, CA, USA) to exclude dead cells.

MTT assay
Effects of over-expression and inhibition of MKRN2 on proliferation of K562 cells were assessed by the methabenzthiazuron (MTT) method. Cells (5610 4 per well) were seeded in duplicates onto a 24-well plate (Corning, NY, USA) and incubated with 100 mL MTT (5 mg/mL; Invitrogen) for 30 min at 37uC. The insoluble violet formazan crystals and cells were collected by centrifugation at 18,3006g for 10 min and dissolved in 100 mL dimethylsulphoxide (DMSO, Invitrogen). Absorbance was read at 570 nm. Duplicate measurements were determined in 3 independent experiments and expressed as percentage of the control.

Reverse transcription and qPCR
Total RNA was extracted from cell cultures (1610 6 /samples), peripheral blood MNC or bone marrow samples using Trizol reagent (Invitrogen). cDNA was synthesized from 1 mg of total RNA using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA). qPCR analysis was carried out using human specific Taqman Gene Expression Assays (Applied Biosystems). These primer and probe sets (MKRN2, Hs00274055_m1 and RAF1, Hs00234119_m1) have been recommended for specific gene expression experiments because they detect the maximum number of transcripts for target genes. Results were expressed as relative to glyceraldehyde-3-phosphate dehydrogenase (GAPDH).

Statistical analysis
The significance of growth or inhibitory effects exerted by overexpression or shRNA suppression of MKRN2 in CD34 + cells and K562 cell line were determined by the paired-samples t test. The differences in MKRN2 and RAF-1 mRNA expression levels between normal and malignant hematopoietic cells, and leukemic cell lines were determined by independent samples t test.
Correlations of MKRN2 and RAF-1 in primary leukemic samples and normal bone marrow samples were analyzed by the Pearson Correlation test. All analyses were performed using SPSS for Windows 17 software (SPSS, Chicago, IL, USA). A P value of #0.05 was considered significant. Results are expressed as mean 6 standard error of the mean (SEM).

mRNA expression of MKRN2 in normal and malignant hematopoietic cells
By qPCR analysis, we observed ubiquitous expressions of MKRN2 and RAF1 in primary hematopoietic cells including adult MNC, neutrophils, cord blood MNC, enriched CD34 + cells, and human embryonic stem cell lines H9 and H14 (n = 2-6) (Fig 1).
The expression levels are represented as relative to GAPDH (Yaxis, Fig 1-3). We also demonstrated expressions of MKRN2 and RAF1 in leukemia cell lines of B-ALL, T-ALL, AML, CML, NK and MK lineages (n = 2-3) (Fig 2). In primary leukemia samples obtained from BM of patients, we showed positive expressions of MKRN2 and RAF1 in B-ALL Philadelphia chromosome (BCR/ ABL or Ph) positive and negative, T-ALL, AML and CML samples (n = 5-22) (Fig 3). The expression levels of MKRN2 were generally higher in leukemia samples (P,0.05 in Ph-B-ALL, Ph+B-ALL, T-ALL and AML samples) compared with those in age-matched normal BM cells (n = 9), whilst RAF1 was higher in Ph-B-ALL and  Over-expression of MKRN2 in cord blood CD34 + cells Lentiviral transduction of MKRN2 resulted in 1.1460.09 fold change of MKRN2 mRNA and 0.9360.27 fold change of RAF1 in total CD34 + cells, relative to their respective levels in control cells containing the empty vector (n = 3). There were trends of increased cell expansion (Fig 4) and the number of multilineage stem/progenitor cells (Fig 5) after ex vivo culture of transfected cell in the presence of cytokine supplement. However, the differences were not statistically significant.

ShRNA-silencing of MKRN2 in cord blood CD34 + cells
ShRNA inhibition of MKRN2 resulted in reduction of MKRN2 expression (0.6660.12 fold vs. pGFP-V-RS; and 0.7960.07 fold vs. pGFP-V-RS-NE) (Fig 6). RAF1 expression in MKRN2-silenced cells was 1.2860.31 fold compared with that in control pGFP-V-RS cells. CD34 + cells expressing pGFP-V-RS-MKRN2 had significantly lower proliferation capacity as shown in day 7 culture (P = 0.005) and a trend of reduced expansion at day 2 and day 8 cultures (n = 3), when compared with either non-effective sh cassette (pGFP-V-RS-NE) or empty vector (pGFP-V-RS) control cultures. The early stem/progenitor cells (CFU-GEMM) were also decreased in the pGFP-V-RS-MKRN2 nucleofected cell expansion culture (Fig 7).  (Fig D in File S1). MTT assay of pLEF1a-IG-MKRN2 transduced cells showed a significantly increased proliferation in culture, compared with that of the pLEF1-IG control cells (P = 0.05) (Fig 8).

Discussion
Our data provided the first evidence on the ubiquitous expression of MKRN2 in multi-lineage normal hematopoietic and leukemic cells, as well as its function on promoting CD34 + and K562 cell proliferation. In spite of the known conservation of the MKRN2 gene through evolution, little has been reported on its role in any organism other than the anti-neurogenic activity in Xenopus laevis [3,4]. Using shRNA silencing, we demonstrated the activity of MKRN2 on promotion of CD34 + cell expansion to early progenitor cells, indicating its role on normal hematopoiesis. However, over-expression of MKRN2 in CD34 + cells did not  significantly affect cell expansion and lineage development, possibly because endogenous levels of MKRN2 protein were sufficient for its cell promoting functions. Upregulated expressions of MKRN2 in primary leukemia cells prompted us to further investigate the effects of forced expression and silencing of MKRN2 in the leukemic cell line K562. Again, we observed the stimulating activity of over-expressing MKRN2 on K562 proliferation. In contrast to CD34 + cells, sh-silencing of MKRN2 in K562 did not affect cell proliferation, indicating possible differences between the regulatory mechanism of MKRN2 in CD34 + cells and leukemic cell line K562. Further evaluation of MKRN2 gene manipulation on cell cycle regulation might reveal its specific mechanism on hematopoietic cell proliferation.
Due to the common sequence between MKRN2 and RAF1 in the antisense orientation, we suspected existence of a mutual regulatory mechanism between the two genes [11,12]. RAF1, a protein closely associated with the RAP1, RAS, ERK and AKT pathways, plays multiple roles in hematopoietic cells [13]. It is required for growth factor-induced proliferation of normal hematopoietic and leukemic cells [14]. RAF1 is also implicated in drug resistance of BCR/ABL expressing leukemic cells [15]. In normal and leukemic cells, however, we only observed ubiquitous expressions of MKRN2 and RAF1. They did not exhibit any convincingly significant correlation in their expression patterns. It is anticipated that a larger sample size of each leukemia subtype would be required to accurately address the relationship between MKRN2 and RAF1, as well as between specific translocations such as BCR-ABL.
To our knowledge, there have been very few reports on the involvement of MKRN2 in malignancy, except some microarray  screening data on papillary thyroid cancer [16]. MKRN1, the most studied member of the MKRN family has been shown to participate in a variety of mechanisms such as RNA-II-dependent transcription [17], Oct-4 signaling in mouse embryonic stem cells [18], telomere length homeostasis in cancer cell lines [19,20], polycystic kidney [21], ubiquitinase activity [22], and p14ARFassociated cellular senescence and gastric tumorigenesis [23]. Based on the ubiquitous expression and proliferative promoting activity of MKRN2 in the various developmental windows of hematopoiesis, we suggest that MKRN2 may play a house-keeping role on normal hematopoiesis. Our study has provided evidence that MKRN2 might also be involved in the proliferation of human leukemic cells. Further knowledge on MKRN2 interaction with known proto-oncogenes and involvement in leukemogenesis may lead to development of alternative treatment for the malignancy.

Supporting Information
File S1 Figure A: MKRN2 construct for lentiviral transduction. Figure B: Correlation of MKRN2 and RAF1 Expression in Leukemia Samples. Expression levels of MKRN2 and RAF1 mRNA, relative to GAPDH, in bone marrow cells collected from leukemic patients (Ph-B-ALL, n = 8; Ph+B-ALL, n = 7; T-ALL, n = 5; AML, n = 22 and CML, n = 11) and age-matched normal bone marrow donors (n = 9) were measured by qPCR and analyzed by Pearson correlation test. A positive correlation (P = 0.042) was observed in Ph+B-ALL samples. However, the correlation became insignificant when the one sample with extremely high expressions of both MKRN2 and RAF1 was excluded from analysis. Ph = Philadelphia chromosome or BCR/ABL translocation. Figure C: Expression of MKRN2 and RAF1 in CML patients with Major or Minor BCR/ABL. Expression levels of MKRN2 and RAF1 mRNA, relative to GAPDH, in bone marrow cells collected from CML BCR/ABL Major (n = 8) and Minor (n = 3) leukemic patients were measured by qPCR. There were no significant differences between the mRNA expression of either genes in the 2 subgroups of CML patients. Ph = Philadelphia chromosome or BCR/ABL translocation. Figure