The Generation of Programmable Cells of Monocytic Origin Involves Partial Repression of Monocyte/Macrophage Markers and Reactivation of Pluripotency Genes

Abstract

We have recently demonstrated that peripheral blood monocytes can be differentiated in vitro into hepatocyte-like cells using appropriate differentiation media. Phenotype conversion required prior in vitro culture in the presence of M-CSF, IL-3, and human serum, during which the cells acquired a state of plasticity, so were termed “programmable cells of monocytic origin” (PCMO). Here, we have further characterized the process of PCMO generation with respect to markers of monocyte-to-macrophage transition and pluripotency. During a 6-day culture period, various monocyte/macrophage differentiation markers were down-regulated being indicative of a process of partial dedifferentiation. Dedifferentiation and hepatic redifferentiation also proceeded in highly purified monocyte preparations, albeit with different kinetics, suggesting that the presence of nonmonocytes, or soluble factors derived from them, is not essential in order for monocytes to acquire a multipotent state. PCMOs expressed various markers of human embryonic stem cells with early induction of NANOG and OCT4. Expression of the pluripotency-associated OCT4A isoform was paralleled by a global rise in histone H3 methylation on Lys-4, a marker of active chromatin, and coincided with peak sensitivity to tissue-specific differentiation. These results show that peripheral blood monocytes can be induced in vitro to transiently acquire stem cell-like properties and concomitantly a state of increased differentiation potential toward the hepatocytic phenotype.

Introduction

A dult stem cells represent a promising alternative to embryonic stem cells (ESCs) in regenerative medicine and in tissue engineering [1]. Although not considered a classical adult stem cell, the peripheral blood monocyte is an extraordinarily versatile progenitor cell, giving rise to very diverse cell types, which include long-lived, self-renewing cell populations. The monocyte ultimately derives from the hematopoietic stem cell (HSC), which is the precursor of the common myeloid progenitor (CMP). From the CMP arises the granulocyte/monocyte progenitor (GMP) and it is from these precursor populations that monoblasts arise. Monoblasts are the earliest form committed to becoming monocytes and having differentiated by stages to monocytes, their progeny emigrate from the bone marrow into the peripheral blood. Peripheral blood monocytes, when appropriately stimulated, will traffic to sites of inflammation and extravasate into the tissues, acquiring the characteristics of an activated macrophage. Alternatively, when not recruited to inflammatory lesions, monocytes are able to undergo a time-dependent maturation into several classes of tissue-resident macrophages, which include resting tissue macrophages, Kupffer cells, Langerhans cells of the skin, dendritic cells, microglia, osteoclasts, and endothelial cells (reviewed in Ref. [2]).

Recently, we have developed a protocol to induce from monocytes by in vitro culture an apparently more plastic derivative, which we named “programmable cells of monocytic origin” (PCMO). These cells following a 6-day treatment with M-CSF, IL-3, and human serum (designated “PCMO culture”) could be induced upon exposure to appropriate induction media to differentiate into cells with certain endothelial characteristics [3], chondrocytes [4], and insulin-expressing cells [5]. The latter cells up-regulate not only the insulin and glucagon genes but also transcription factors involved in pancreatic β-cell differentiation [5]. Recently, we focused our interest on PCMO-derived hepatocyte-like cells (NeoHeps), which express a variety of hepatocyte markers and exhibit hepatocyte-specific metabolic functions [5 –7], making these cells an attractive alternative to primary human hepatocytes for studying drug metabolism in vitro [7]. Moreover, NeoHeps improved survival in a rat model of acute liver failure [8] and monocyte-derived hepatocyte-like cells even showed promise in the treatment of HBV-related decompensated liver cirrhosis [3,9]. The crucial point, from a bio(techno)logical perspective, is that monocytes are capable of coordinated expression of the gene sets required for these nonhematopoietic cell-like activities to occur.

Various mechanisms have been implicated in the acquisition of plasticity/enhanced differentiation potential of adult somatic cells including myelomonocytic cells, such as trans-differentiation, dedifferentiation, or cell fusion [10]. During dedifferentiation, cells silence tissue/cell-type-specific genes and eventually reacquire more primitive features, such as mitotic activity and expression of markers of self-renewal and multi/pluripotency. This would be consistent with previous observations that PCMOs down-regulated PRDMI, the human homolog of murine BLIMP-1, and ICSBP, 2 genes encoding monopoietic transcription factors [5]. However, no information is yet available for other nuclear factors of monocyte → macrophage differentiation (eg, Klf4) or markers associated with specialized functions, for example those involved in sensing and killing of invading microorganisms such as Toll-like receptors (TLRs) and NAD(P)H oxidase, respectively. A more general loss of the monocyte/macrophage phenotype would be in favor of a dedifferentiation process and would lend support to our contention that PCMOs represent cells that have reverted to a more primitive progenitor with less restricted differentiation potential. Hence, in the first part of this study we examine whether altering specific parameters, such as duration of PCMO culture or cellular composition of the monocyte preparation, impact on acquisition of the multipotent state by PCMOs.

Despite the biological importance of the broad differentiation potential of monocytes, surprisingly little is known about the mechanisms that allow them to maintain an uncommitted precursor state. It is reasonable, then, to question whether the mechanisms underlying the phenotypic plasticity of embryonic and adult stem cells might also operate in monocytes subjected to PCMO culture conditions. In this respect, we provide evidence that PCMOs share in common several markers of ESCs such as OCT4 (also termed POU5f) and NANOG, both of which can function as activators of self-renewal and pluripotency genes and as repressors of lineage commitment genes. Specifically, we demonstrate that PCMOs reactivate the endogenous OCT4 and NANOG genes and undergo histone modifications known to be associated with their transcriptional activation. These genetic and biochemical events correlated well with respect to timing and duration with the transiently increased differentiation potential of the cells. We surmise that the generation of PCMOs involves both partial dedifferentiation and the acquisition of stem cell-like properties, providing a molecular explanation for the previously observed phenotypic conversion of peripheral blood monocytes into hepatocyte-like cells.

Materials and Methods

Isolation, purification, and in vitro culture of monocytes

PCMOs were generated from human peripheral blood monocytes following the protocol of Ruhnke and colleagues [5]. In brief, mononuclear cells from either heparinized blood, buffy coats, or LRS chambers were isolated by density gradient centrifugation (Ficoll-Paque; Amersham Pharmacia Biotech AB, Uppsala, Sweden). Cells were cultured in either 24-well or 6-well plates (Cell+, Sarstedt, Nümbrecht, Germany) for various lengths of time in PCMO medium (RPMI 1640 medium (Invitrogen, Karlsruhe, Germany), supplemented with 5 ng/mL final concentration of M-CSF and 0.4 ng/mL final concentration of IL-3 (both from R&D Systems, Wiesbaden, Germany), 90 μM 2-mercaptoethanol, and 10% human AB serum (Lonza, Verbier, Belgium). One hour after plating, cultures were gently washed to enrich for adherent cells and fresh medium was added to the adherent cell layer resulting in enrichment of 60%–70% CD14⁺ cells. These cultures are referred to as Ficoll-Paque density gradient centrifugation (FC)-purified. Alternatively, monocytes were purified by elutriation as described [11], based on a previously published method [12], resulting in a purity of 95%–97% CD14⁺ monocytes. Purified cells were cultured for various times (1 h, 2, 4, or 6 days) in PCMO medium, which was replaced in all cultures every other day. Subsequently, hepatocytic differentiation was initiated by replacing the PCMO medium with RPMI medium containing 10% fetal calf serum (Lonza), 3 ng/mL final concentration of FGF-4, and 20 ng/mL final concentration of hepatocyte growth factor (HGF; both from R&D Systems). The addition of HGF, which was not part of the original medium formulation [5,6], further enhanced albumin and Factor VIII expression (A.N. and H.U., unpublished data).

Flow cytometry analysis

Flow cytometry analysis of freshly isolated and cultured monocyte preparations was carried out as described previously [5] with minor modifications. The following antibodies were employed: APC-conjugated mouse antihuman CD14 (clone M5E2; BD Biosciences, San Jose, CA), fluorescein isothiocyanate (FITC)-conjugated antihuman TLR2 (clone TL2.1), and phycoerythrin (PE)-conjugated antihuman TLR4 (clone HTA125, both from eBioscience via NatuTec, Frankfurt, Germany). Only viable cells negative for 7-amino-actinomycin D (7-AAD) were analyzed.

RNA isolation and RT-PCR

RNA was isolated with PeqGold (peqLab, Heidelberg, Germany) according to the manufacturer's protocol. Due to the lack of introns in both OCT4 and NANOG and the presence of several pseudogenes in the human genome, total monocyte RNA preparations were always treated with DNase I (Invitrogen) prior to reverse transcription. Primers were generally chosen to span exon–intron boundaries and, in the case of NANOG, to be unable to recognize the mRNA sequences encoded by the 3 expressed pseudogenes [13]. As positive control, we employed total RNA isolated from the human ESC line SA002.5 (kindly donated by Dr. P. Björquist, CellArtis, Stockholm, Sweden). Standard end-point PCR was performed on a My-Cycler (Bio-Rad, München, Germany) with Taq polymerase and supplied buffers (Invitrogen) and a hot start-touch-down program described earlier [6]. In brief, primers and cDNA template were premixed in a volume of 2.5 μL and added to the remaining 22.5 μL of PCR preheated to 80°C. Denaturation, annealing, and extension were performed at 30 s each. The initial annealing temperature of 61°C was lowered by 2°C every second cycle until a temperature of 51°C was reached at which 30 cycles were performed. PCR products were analyzed by electrophoresis through 2% agarose gels and viewed under UV light after ethidium bromide staining. Quantitative real-time RT-PCR analysis (qPCR) was carried out on an I-Cycler with iQ SYBR Green Supermix (Bio-Rad). Denaturation, annealing, and extension were performed at 30 s each. Forward and reverse primers were designed to have approximately the same T _m and annealing temperatures were chosen to be T _m −5°C. Calculation of results was performed with the ΔΔC_T method following normalization of the data from the gene of interest with equivalent data for the 2 housekeeping genes, TATA box-binding protein (TBP) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH). For oligonucleotides used as PCR primers (see Table 1).The primer sequences for ICSBP, PRDMIβ, PU-1, GAPDH [5] and Nox1–5, p22^phox, p47^phox, and p67^phox [14] have been given earlier.

Table 1.

List of PCR Primers

Primer name	Sequence (5′→′)	GenBank accession
ABCG2-for	GTTTATCCGTGGTGTGTCTGG	NM_004827
ABCG2-rev	CTGAGCTATAGAGGCCTGGG	NM_004827
Albumin-for	CTCTTTAGCTCGGCTTATTCC	NM_000477
Albumin-rev	TCAACCTCTGGTCTCACCAAT	NM_000477
BMP-2-for	AGACCTGTATCGCAGGCACT	NM_001200
BMP-2-rev	AAACTCCTCCGTGGGGATAG	NM_001200
CD105-for	CAGCTGTGGCATGCAGGTGTC	NM_000118
CD105-rev	GTACTCACGCGTGTGCGAGTAG	NM_000118
Connexin-43-for	TACCATGCGACCAGTGGTGCGCT	NM_000165
Connexin-43-rev	GAATTCTGGTTATCATCGGGGAA	NM_000165
Cripto/TDGF1-for	CAGCACAGTAAGGAGCTAAACA	NM_003212
Cripto/TDGF1-rev	CAGTTCCGTCCGTAGAAGGAG	NM_003212
DNMT3b-for	AATGTGAATCCAGCCAGGAAAGGC	NM_006892
DNMT3b-rev	ACTGGATTACACTCCAGGAACCGT	NM_006892
E-cadherin-for	TCTTCCCCGCCCTGCCAATC	Z13009
E-cadherin-rev	GCCTCTCTCGAGTCCCCTAG	Z13009
Factor VIII-for	CTTCATGTCGATGGAAAACC	NM_000132
Factor VIII-rev	TGTGGTGGCATTAAATTGCT	NM_000132
GABRB3-for	CAAGCTGTTGAAAGGCTACGA	NM_000814
GABRB3-rev	GCGATGTCGATGTTCATCCC	NM_000814
GAPDH-for	TTGCCATCAATGACCCCTTCA	NM_002046
GAPDH-rev	CGCCCCACTTGATTTTGGA	NM_002046
Gdf3-for	AGACTTATGCTACGTAAAGGAGCT	NM_020634
Gdf3-rev	CTTTGATGGCAGACAGGTTAAAGTA	NM_020634
Id2-for	GACCCGATGAGCCTGCTATAC	NM_002166
Id2-rev	AATAGTGGGATGCGAGTCCAG	NM_002166
KDR-for	AAGGTGACAGGAAAAGACGAACT	NM_002253
KDR-rev	TCCCCTCCATTGGCCCGCTTAAC	NM_002253
KLF4-for	GCCACCCACACTTGTGATTA	NM_004235
KLF4-rev	CGTCCCAGTCACAGTGGTAA	NM_004235
Lefty B-for	TGCTACAGGTGTCGGTGCAGAGG	NM_020997
Lefty B-rev	AGAAACGGCCACTTGAAGGCCAGG	NM_020997
c-Myc-for	ATGCCCCTCAACGTTAGCTTCAC	NM_002467
c-Myc-rev	TTTAGCTCGTTCCTCCTCTGGC	NM_002467
Nanog-for	GATTTGTGGGCCTGAAGAAAACT	NM_024865
Nanog-rev	AGGAGAGACAGTCTCCGTGTGAG	NM_024865
NCAM-for	AGGAGACAGAAACGAAGCCA	NM_000615
NCAM-rev	GGTGTTGGAAATGCTCTGGT	NM_000615
Nodal-for	CGACCAACCATGCATACATCC	NM_018055
Nodal-rev	GTGACTTCATCCCACCTCCAA	NM_018055
Oct4-for	GACAACAATGAAAATCTTCAGGAGATATG	NM_002701
Oct4-for216	CTCTGAGGTGTGGGGGATTC	NM_002701
Oct4-rev	TTCTGGCGCCGGTTACAGAACC	NM_002701
Rex1-for	CAGATCCTAAACAGCTCGCAG	NM_174900
Rex1-rev	CCCGTGTGGATGCGCACGT	NM_174900
Sox2-for	AGTCTCCAAGCGACGAAAAA	NM_003106
Sox2-rev	GGAAAGTTGGGATCGAACAA	NM_003106
STELLA-for	GTTACTGGGCGGAGTTCGTA	AY317075
STELLA-rev	TGAAGTGGCTTGGTGTCTTG	AY317075
TBP-for	GCTGGCCCATAGTGATCTTT	M55654
TBP-rev	CTTCACACGCCAAGAAACAG	M55654
hTERT-for	CGGAAGAGTGTCTGGAGCAA	NM_198253
hTERT-rev	GGATGAAGCGGAGTCTGGA	NM_198253
TLR2-for	TTGCTCTTTCACTGCTTTCAACT	NM_003264
TLR2-rev	TCCCACTCTCAGGATTTGCAA	NM_003264
TLR4-for	ATGGGGCATATCAGAGCCTA	NM_138554
TLR4-rev	CACAGCCACCAGCTTCTGTA	NM_138554
TLR7-for	TAGGATCACTCCATGCCATC	NM_016562
TLR7-rev	GAGTCCACTAAAGCTTCTGG	NM_016562
TLR9-for	ACTGGCTGTTCCTGAAGTCT	NM_017442
TLR9-rev	CAGTGAGTTGTTGTACTTGAG	NM_017442
UTF1-for	ACCAGCTGCTGACCTTGAAC	NM_003577
UTF1-rev	TTGAACGTACCCAAGAACGA	NM_003577

Isolation of nuclear protein, immunoblotting, and immunoprecipitation

Preparation of total cellular lysates, SDS-PAGE, and immunoblotting were carried out as described in detail elsewhere [15]. Nuclear proteins from PCMOs, F9 murine embryonal carcinoma, and JEG-3 human choriocarcinoma cells were isolated using a commercially available kit (Thermo Scientific, Rockford, IL) and protein concentrations were determined with the Bradford assay (Thermo Scientific). Proteins were fractionated by SDS-PAGE, blotted onto PVDF membrane, and probed with the following antibodies: p47^phox (Santa Cruz Biotechnology, Santa Cruz, CA), c-Src (GD11; Upstate Biotechnology, Waltham, MA), β-actin, α-tubulin (both from Sigma, St. Louis, MO), E-cadherin (BD Biosciences), Oct4 (Santa Cruz Biotechnology, # sc-5279), Nanog (Abcam, Cambridge, UK), histone H3 (trimethyl K4), histone H3, and histone H4 (all from Biozol, Eching, Germany). Oct4 and heat-shock protein (HSP) 90 were also immunoprecipitated from PCMO total protein lysates with monoclonal Oct4 and HSP90α/β (H-114) antibodies, respectively, and Protein A/G-Plus Agarose (all from Santa Cruz Biotechnology) according to the protocol of the supplier. Following SDS-PAGE and blotting, immunoprecipitated OCT4 and HSP90 proteins in PCMOs and in F9 cells (purchased as positive control lysate from Santa Cruz Biotechnology) were detected using the above mentioned antibodies.

Statistics

Data from qPCR experiments represent the mean ± SD after normalization with the housekeeping genes GAPDH and TBP and were generally derived from 3 wells processed in parallel. Statistical significance was calculated using the unpaired Student's t-test. Data were considered significant at P < 0.05.

Results

The monocyte–PCMO transition is characterized by changes in marker expression for monocyte/macrophage function and cell adhesion

We have previously observed that in response to our specific culture conditions monocytes silence the monopoietic marker genes PRDMI and ICSBP [5]. To reveal whether down-regulation of these genes was part of a more general process of dedifferentiation, we monitored various proteins involved in the sensing and killing of bacteria by monocytes/macrophages, such as CD14, Toll-like receptors (TLRs) 2 and 4, as well as NOX4 and p47^phox both subunits of the reactive oxygen-producing enzyme NAD(P)H oxidase. Flow cytometry analysis of cell surface expression of CD14, TLR2, and TLR4 in freshly isolated monocytes and PCMOs of various stages of maturity (n = 4) indicated that the percentage of CD14⁺ cells decreased to 46.6% (range: 30.3%–61.2%) on Day 6 of culture, while the corresponding values for TLR2 and TLR4 were 4.7% (range: 1.6%–8.4%) and 14.1% (range: 4.6%–26.6%), respectively (Fig. 1A, see also means ± SD for each donor). A time-course analysis revealed that the decrease in CD14 and TLR2 expression commenced between Days 1 and 4 of culture while that of TLR4 was much more rapid and was virtually complete after 1 day (Fig. 1A). Other markers (CD86, HLA-DR) remained constant over the 6-day culture period (data not shown). To evaluate whether down-regulation of the TLRs is controlled at the transcriptional level, we employed qPCR analysis. Interestingly, expression of TLR2 (Fig. 1B), TLR4 (see Fig. 4A), 2 other TLRs, TLR7 (recognizing viral RNA) and TLR9 (recognizing bacterial DNA), and NOX4 (Fig. 1B) was dramatically decreased on Day 6 of culture. In contrast, expression of p47^phox remained unaltered at the mRNA level (data not shown), but appeared to be down-regulated at the protein level starting from Day 2 in PCMO medium (Fig. 1C), consistent with earlier reports that p47^phox expression is primarily controlled posttranslationally [16]. The switch from suspended to adherent growth, which was associated with changes in cell morphology [5], might also affect proteins involved in cell adhesion and cytoskeletal regulation. In fact, we observed a steady increase in p60^Src, β-actin, and E-cadherin (Fig. 1C), while α-tubulin remained fairly constant over the 6-day culture period (Fig. 1C). Together, these data suggest that monocytes during their conversion to PCMOs undergo partial dedifferentiation.

FIG. 1.

Analysis of monocyte/macrophage markers during programmable cells of monocytic origin (PCMO) culture. Peripheral blood monocytes prepared by Ficoll-Paque density gradient centrifugation (FC) were allowed to adhere for 1 h, or cultured for the indicated times, in PCMO medium prior to harvest. (A) Analysis of cell surface expression of CD14, TLR2, and TLR4 in 4 different donors by flow cytometry. Shown is a representative experiment out of 3 experiments each with 4 donors. Means ± SD for each donor were calculated from triplicate samples. (B) Analysis of expression of TLR2, TLR7, TLR9, and NOX4 in 2 different donors by means of qPCR. Data represent the mean ± SD from 3 wells processed in parallel following normalization with the housekeeping genes glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and TATA box-binding protein (TBP). (C) Time-course analysis of the indicated proteins during PCMO culture. Total cellular protein was prepared from the cells at various time points and analyzed by immunoblotting with specific antibodies. Detection of α-tubulin served as loading control.

Optimal hepatocytic differentiation of PCMOs varies with time in culture

In vitro studies with mouse monocytes have shown that the intrinsic sensitivity toward exogenous differentiation clues to a specific phenotype varies with time in culture, with the differentiation potential to macrophages being the default state [17]. To test whether this also applies to human monocytes, we cultured FC-purified monocytes for up to 6 days in PCMO medium, followed by a 14-day exposure to hepatocyte differentiation medium (Fig. 2A) and monitored the expression of hepatocyte-defining markers as earlier. As shown in Figure 2B, the susceptibility of the cells to hepatocytic differentiation indeed was time-dependent with all 3 markers displaying the highest expression when the cells were kept for 4 days rather than for 6 days in PCMO medium. From these data, we conclude that the susceptibility of cultured monocytes to tissue-specific differentiation fluctuates with time in culture.

FIG. 2.

Optimal hepatocytic differentiation of programmable cells of monocytic origins (PCMOs) varies with time in culture. (A) Treatment scheme for determining the highest susceptibility to hepatocyte differentiation. Monocytes were cultured for various times, as indicated, in standard PCMO medium (bold lines) and, subsequently, in hepatocyte-conditioning medium for 14 days to induce NeoHeps (thin lines). (B) Analysis of the relative degree of hepatocytic differentiation by means of qPCR for the indicated genes. Data are from 2 representative donors and are depicted as the mean ± SD of 3 wells after normalization to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and TATA box-binding protein (TBP). Asterisks indicate significance versus control (= cells cultured for 1 h). *p < 0.05.

Dedifferentiation and hepatic redifferentiation also proceed in highly purified monocyte preparations

Monocytes isolated by FC still contain ∼30%–40% CD14⁻ cells, primarily T- and B cells, after adhesion, and this number is further reduced to ∼10%–20% by Day 6 (Ref. [5] and unpublished data). In order to test whether the generation of PCMOs from CD14⁺ monocytes was dependent on the presence of these “contaminating” cells (acting on PCMOs through either cell–cell contacts or soluble factors), we subjected highly purified cultures (>95%) of CD14⁺ monocytes derived by elutriation to the same dedifferentiation–NeoHep differentiation protocol. We observed down-regulation of PRDMI (Fig. 3A) and ICSBP, but constant expression of the promyeloid Ets-family transcription factor Pu.1 (data not shown) as previously described in FC-purified preparations [5]. TLR expression, as exemplified here by TLR4, was rapidly down-regulated (Fig. 3A). Recently, Krüppel-like factor 4 (Klf4) was identified to be expressed in a monocyte-restricted and stage-specific pattern during myelopoiesis and to promote monocyte differentiation [18]. Klf4 was rapidly silenced following exposure of monocytes to PCMO culture conditions and transcript levels remained low until Day 6 (Fig. 3A) after which they became undetectable. The proliferation-associated c-MYC gene was transiently activated with peak expression on Day 2 (Fig. 3A). The reciprocal expression of PRDMI and MYC during the first 2 days of PCMO culture may indicate reactivation of MYC as consequence of its derepression from PRDMI, since ectopic Blimp-1, the murine homolog of human PRDMI, repressed myc expression and inhibited cell growth [19]. Expression of ID2, which is induced by serum and acts as an inhibitor of cellular differentiation [20], peaked on Day 2 and subsequently declined (Fig. 3A). Expression of KDR, a marker of endothelial progenitor cells, was generally low, but appeared to be regulated peaking at the 48-h time point (Fig. 3A). Transcript levels of CD105/endoglin, a marker of both ESCs and endothelial progenitors, were elevated between 24 and 72 h of culture and on Day 6 (Fig. 3A). In a second series of experiments, we successively cultured elutriated monocytes (5 different preparations) for up to 6 days in PCMO medium prior to NeoHep differentiation and subsequently determined albumin expression in these cells. One major “hotspot” of differentiation sensitivity was noted early on Day 2 of culture (Fig. 3B). A similar profile was found for the Factor VIII gene (data not shown). These data indicate that monocyte dedifferentiation is independent from the presence of CD14⁻ cells and that under these conditions the cells more rapidly acquire a plastic state.

FIG. 3.

Dedifferentiation and hepatic redifferentiation proceeds in highly purified monocyte preparations. (A) Time course of monocyte differentiation marker expression (as indicated) during standard programmable cells of monocytic origin (PCMO) culture of monocytes purified by elutriation. Either data are the mean of 3 different donors (KLF4) or the data represent from 1 representative donor out of 5 different donors analyzed (PRDMI, MYC, ID2, KDR, CD105). In this case, the mean ± SD of triplicate wells is given following normalization with glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and TATA box-binding protein (TBP). Asterisks indicate significance versus control (= noncultured cells for Klf4, Id2, CD105, and cells cultured for 1 h in PCMO medium for PrdmI, Myc, TLR4, KDR). (B) Determination of the time point of the greatest differentiation potential in elutriation-purified monocytes. Following culture for various times (as indicated) in PCMO medium, monocytes were incubated in hepatocyte differentiation medium for 14 days according to the scheme in A, and subjected to qPCR for albumin. Data are from 2 representative donors and are depicted as the mean ± SD of 3 wells processed in parallel after normalization of albumin values with those for GAPDH and TBP. Asterisks indicate significance versus control (= cells cultured for 1 day in PCMO medium). *p < 0.05.

PCMOs express hES markers

The time-dependent pattern of sensitivity toward hepatocytic differentiation likely reflects underlying changes in multipotency and may be regulated by genes known to control self-renewal and pluripotency in ESCs. RT-PCR-based screening of FC-purified and elutriated monocytes revealed that on Day 6 of culture cells were consistently positive for OCT4, REX1, and NANOG, while SOX2 was negative in all (5/5) preparations tested (Fig. 4A). The failure to detect SOX2 confirmed the absence of other stem cell types such as HSCs, mesenchymal stem cells (MSCs), and neural stem cells in the PCMO cultures. Due to the lack of introns in OCT4 and NANOG and the presence of pseudogenes in the human genome, we have carefully evaluated that the signal was derived from mRNA rather than genomic DNA contamination and by using primers that do not recognize pseudogenes [13]. OCT4 expression was also verified at the protein level following immunoprecipitation and subsequent immunoblotting of total cellular proteins with a monoclonal antibody recognizing a single epitope localized at amino acids 1–134 (Fig. 4B), thus precluding detection of the OCT4B isoform ([21] and see below). Prior enrichment of nuclear and cytoplasmic proteins from PCMOs followed by direct immunoblotting revealed that both OCT4A and NANOG proteins were more abundant in the nuclear fraction (Fig. 4C).

FIG. 4.

Expression of pluripotency markers in programmable cells of monocytic origins (PCMOs). (A) Day 6 PCMOs from Ficoll-Paque density gradient centrifugation (FC)-purified monocytes and human embryonic stem cells (ESCs) (SA002.5 cell line) were subjected to RNA isolation and DNase I digestion. Following cDNA synthesis, standard end-point PCR reactions were performed with primers for the indicated genes and amplified products resolved by gel electrophoresis and visualized by ethidium bromide staining. The arrows below the images point to the lanes with the mRNA-specific signals. RT, reverse transcriptase. (B) Sequential immunoprecipitation of OCT4 and HSP90 from 50 μg of Day 3 PCMO lysates from the indicated donors, and 50 μg of HepG2 hepatocellular carcinoma cell lysate as negative control, followed by immunoblot analysis using the same antibodies. A commercially available lysate from F9 embryonic carcinoma cells (10 μL) was loaded as positive control. (C) Direct immunoblot analysis of Oct4 (upper panel) and Nanog (lower panel) in cytoplasmic (C) and nuclear (N) proteins from FC-purified Day 3 PCMOs (25 μg/lane). F9 cells (10 μg/lane) and JEG-3 cells (25 μg/lane each) were used as positive control for Oct4A and Nanog [48], respectively. Successful enrichment of nuclear and cytoplasmic proteins was verified with antibodies to histone H4 and β-actin, respectively. Equal protein loading between lanes with cytoplasmic and nuclear proteins of the same sample was confirmed by staining the blots with Ponceau S (not shown). Note that for F9 cells less protein was loaded to avoid overexposure of the bands due to much higher Oct4A expression in F9 cells compared with PCMOs. (D) Standard end-point RT-PCR detection of the indicated genes in Day 6 PCMOs prepared from elutriation-purified monocytes from 5 different donors.

Other genes besides c-MYC and KLF4 implicated in maintenance of self-renewal and pluripotency and to cluster with hESCs [22] and induced pluripotent stem (iPS) cells were expressed in PCMOs generated from elutriated monocytes and include (positive/total number of donors analyzed): reduced expression 1 (REX1) (3/5), growth and differentiation factor 3 GDF3 (3/5), DPPA3/STELLA/PGC7 (4/5) (not shown), ABCG2 (5/5), Connexin-43 (5/5), NCAM (5/5), DNMT3b (5/5), UTF1 (3/5), BMP2 (4/5), CRIPTO/TDGF1 (5/5), E-CADHERIN (5/5), and CD105 (5/5) (Fig. 4D). QPCR analysis showed that these genes were expressed at a much lower level than in NeoHeps, while others like GABRB3, NODAL, LEFTY B, and telomerase reverse transcriptase (hTERT) were negative in both PCMOs and NeoHeps from all 5 preparations analyzed (data not shown).

Reactivation of pluripotency markers and global histone H3 Lys-4 methylation during PCMO generation

Two different mRNAs are generated from the OCT4 gene by alternative splicing and encode 2 isoforms designated OCT4A and OCT4B [23]. Only OCT4A is associated with pluripotency, but not the related OCT4B, which lacks sequences encoded by exon 1 [21,24]. To distinguish between both isoforms, we monitored OCT4 expression at various time points during the monocyte → PCMO transition in elutriated monocytes using 2 different upstream primers. One recognized both transcripts (OCT4A + OCTB), while the other one bound to exon 1 and specifically amplified OCT4A. We observed that expression of both total OCT4 and OCT4A peaked on Days 2–3 and subsequently declined (Fig. 5A), and was undetectable in NeoHeps (data not shown). Notably, OCT4A expression correlated temporally with the susceptibility “hotspot” to hepatic differentiation (compare Fig. 3B). Interestingly, NANOG expression in elutriated monocytes followed a similar time course, but was slightly delayed relative to that of OCT4 (Fig. 5B). The data indicate that critical pluripotency regulators (eg, OCT4, NANOG) and possibly other factors of the pluripotency network are reactivated in PCMOs and are likely to control their plastic behavior.

FIG. 5.

Kinetics of OCT4 and NANOG activation and histone H3(K4) methylation in programmable cells of monocytic origins (PCMOs). (A) Complementary DNA from elutriation-purified monocytes cultured for various times in PCMO medium, and from human SA002.5 embryonic stem cells (ESCs) as control, served as template for OCT4 qPCR using primers which distinguish, or not, the OCT4 A and B isoforms. Data represent the mean ± SD from 3 parallel wells after normalization with the corresponding values for glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and TATA box-binding protein (TBP). Note that the bar for ESCs is not in scale. (B) Time course of NANOG expression from the same donors analyzed in A as determined by qPCR. Data represent the normalized mean ± SD of 3 wells processed in parallel. Note that the bar for ESCs is not in scale. (C) Time course of global histone H3 Lys-4 trimethylation (me-H3(K4)) during PCMO generation. Ficoll-Paque density gradient centrifugation (FC)-derived monocytes were incubated in PCMO medium for the indicated lengths of time and harvested for total cellular protein isolation. Protein lysates were fractionated by SDS-PAGE and probed with an antibody against trimethyl-histone H3(K4). Total protein loading was controlled with antibodies against histone H3 and β-actin. Asterisks indicate significance versus control (= cells cultured for 1 day). *p < 0.05.

Histone methylation plays a crucial role in epigenetic regulation of gene expression during mammalian development. In general, transcribed genes are associated with trimethylation at Lys-4 of histone H3 (me-H3(K4)) [25], whereas many silenced genes are associated with H3(K27) trimethylation [26]. In undifferentiated cells, the OCT4 promoter is packaged with nucleosomes that contain markers of active chromatin, namely histone H3 highly acetylated at Lys-9 and Lys-14, and me-H3(K4) [27,28]. In both adherent FC-purified monocytes (Fig. 5C) and in PCMO cultures from elutriated monocytes (data not shown), we noted a transient rise in global me-H3(K4) that closely mirrored the time course of OCT4 transcriptional activity. The methylation of the OCT4 promoter remained high during various stages of PCMO generation, as revealed by bisulfite conversion and pyrosequencing at specific CpG islands in the OCT4 distal enhancer [29] (Supplementary Fig. 1; Supplementary materials are available online at http://www.liebertpub.com/), suggesting that promoter demethylation was not the underlying mechanism of OCT4 induction. Taken together, our data support the view that nuclear reprogramming events underlie the process of PCMO generation.

Discussion

In this study, we have analyzed in more detail the molecular changes that occurred during the conversion of freshly isolated peripheral blood monocytes to PCMOs. We noted rapid down-regulation of TLRs 2, 4, 7, and 9, NOX4, p47^phox, and KLF4, suggesting reduced capacity for microbial sensing/killing, and reduced ability for proinflammatory signaling and differentiation in these cells [30,31]. The role of KLF4 is particularly interesting, since its forced expression in primary CMPs or HSCs induced exclusive monocyte differentiation and activated the CD14 promoter [18]. Hence, its down-regulation in PCMOs may be a prerequisite for dedifferentiation and may account for the partial loss of CD14 expression (see Fig. 1A). However, since KLF4 is also essential for efficient genetic reprogramming in the process of induced pluripotency [32], residual KLF4 along with OCT4 [32] is likely required for PCMOs to acquire a multipotent state. Interestingly, however (ectopic) expression of OCT4 in certain somatic cells has also been associated with active dedifferentiation [33]. The expression patterns of the above-mentioned differentiation markers and of the ID2 and MYC genes (see Fig. 3A) strengthen our hypothesis that the culture conditions applied induce monocytes to acquire a more undifferentiated phenotype.

Studies with ex vivo Flt3 ligand-expanded murine monocyte/macrophage precursors showed that their intrinsic sensitivity toward exogenous in vitro differentiation to a specific phenotype changes with time in culture [17]. This mirrors a sequential commitment to macrophage (M-CSF), osteoclast (M-CSF + RANKL), dendritic cell (GM-CSF + TNF-α), or microglia (glial cell-conditioned medium) differentiation in response to the respective differentiation agents, while the differentiation potential to macrophages was constitutive [17]. This suggests that the macrophage phenotype is the default pathway and that windows for optimal differentiation toward other phenotypes also exist in cultured human monocytes. Indeed, results of this study showed that within the time frame analyzed monocytes exhibited a peak of increased sensitivity to external (hepatocytic) differentiation cues; peak plasticity was recorded on Day 4 in FC-purified monocytes and on Day 2 in elutriated monocytes. Since in FC-derived monocytes the 4-day time point correlated with maximal mitotic activity [5], it is tempting to speculate that proliferation is a prerequisite for the cells to adopt a nonhematopoietic phenotype. Dooner and coworkers indeed demonstrated that marrow cells with the capacity to convert into cells with a pulmonary epithelial phenotype after transplantation showed an increase with cytokine (eg, IL-3)-induced cell cycle transit [34].

As discussed earlier, our culture conditions apparently allowed for limited self-renewal capacity and considerable dedifferentiation of monocytes toward a more primitive progenitor that shares some features with the progenitor of monocytes and granulocytes (GMP) [5]. Interestingly, Charrière and coworkers found similarities between transcriptomic profiles of macrophages and murine undifferentiated ESCs, but not differentiated stem cells [35]. Dedifferentiation combined with proliferation is also the immediate response toward induction of the reprogramming factors OCT4, SOX2, KLF4, and c-MYC in the process of “induced pluripotency” in terminally differentiated cells [36]. This prompted us to hypothesize that PCMO generation involved genetic reprogramming events including reactivation of ESC markers and/or genes regulating pluripotency and self-renewal. In elutriated monocyte preparations we have detected, amongst others, mRNA encoding OCT4, KLF4, MYC, NANOG, and the OCT4 target genes REX1 and UTF1, the notable exception being SOX2. Intriguingly, hotspots of differentiation sensitivity indeed correlated not only with elevated expression of MYC and ID2 (markers of proliferation and suppressed differentiation, respectively), but also with transiently induced OCT4 and NANOG in the cultures (compare Figs. 3 and 5). So far, OCT4 expression in other multipotential subsets of monocytes has only been reported for monocytic endothelial progenitor cells [13] and in PBMCs at both the mRNA and protein levels [37]. However, the tools used by these authors preclude definite conclusions as to whether the A isoform of OCT4 (the only isoform associated with pluripotency [21,24]) has been detected. By using PCR primers for exon 1 and a monoclonal antibody recognizing the exon 1-encoded N-terminal epitope, we were able to demonstrate transcriptional induction of OCT4A during exposure of CD14⁺ cells to M-CSF, IL-3 and serum. The finding that elevated OCT4(A) expression coincided with both the onset of proliferation, up-regulation of the proliferation-associated MYC gene, and the time of peak sensitivity to external hepatocyte differentiation clues clearly argues in favor of OCT4 determining multipotency in PCMOs and we are currently performing functional studies to test this assumption. However, in light of recent doubts on such a role in adult somatic cells [38], alternative explanations have to be considered. One possibility is that OCT4 has a role in driving cell cycle progression (as reported for MSCs [39]) rather than in conferring pluripotency, and that its up-regulation in cycling cells or during specific phases of the cell cycle allows for a higher sensitivity to external differentiation clues merely as a “collateral effect.” It may also be possible that the reactivation of OCT4 in monocytes/PCMOs in response to culture conditions does not reflect a physiological mechanism in place to maintain somatic stem cells in a undifferentiated state in vivo, but may nevertheless result in a therapeutically relevant cell type [38].

PCMOs appear to display several features of partially reprogrammed cell lines [36] such as partial reactivation of genes related to stem cell renewal and maintenance (such as MYC), and some pluripotency genes (OCT4, NANOG), and incomplete repression of lineage-specifying transcription factors (such as PU.1 [5]). A third feature of stable partially reprogrammed cell lines is DNA hypermethylation at pluripotency-related loci. Interestingly, the distal enhancer region of OCT4 remained highly DNA-methylated in PCMOs (see Supplementary Fig. 1). Currently, it is not known for PCMOs whether histone H3 Lys-4 is methylated and H3 Lys-27 is demethylated in the promoter regions of OCT4 and NANOG as in human iPS cells [40]. Based on these observations it may be possible to further enhance genetic reprogramming of PCMOs and eventually induce a pluripotent state through (1) genetic complementation of SOX2, (2) silencing of PU.1, which has recently been shown to repress stem cell capacities in HSCs [41], or DNMT1, or (3) treatment of PCMOs with small-molecule compounds that modify chromatin or enhance the action of pluripotency factors [42].

It is currently unclear whether the fluctuation(s) in OCT4 and NANOG expression reflect transcriptional regulation of these genes in the majority of cells in a given culture, or whether it is the result of preferential expansion of a subpopulation of CD14⁺ cells with constitutively high expression of these markers. The dramatic changes seen in histone modifications during PCMO generation clearly argue in favor of the first scenario. An assembly of available data on cultured human cell populations that originate from circulating monocytes and have the potential to differentiate into nonphagocytes [43] has shown that these monocyte derivatives share the expression of CD14, CD45, and CD34. With respect to expression of CD34 detected by conventional flow cytometry (on a minor fraction [5]) and qPCR, and KDR detected by qPCR (this study), PCMOs resemble the CD14⁺CD34^lowKDR⁺ subset described by Romagnani and colleagues [13]. PCMOs, like pluripotent stem cells (PSCs) [44], require M-CSF, but unlike monocyte-derived multipotential cells (MOMCs) [45], they require neither fibronectin (H. U., unpublished data) nor factors secreted by CD14⁻ cells. A transient pattern of PCMO sensitivity to NeoHep differentiation was observed in both standard FC-purified monocyte cultures (containing a variable number of CD14⁻ cells) and highly pure preparations of CD14⁺ cells derived by elutriation. However, peak sensitivity in crude cultures was delayed by ∼48 h, suggesting that CD14⁻ cells in monocyte/PCMO cultures inhibit the rapid acquisition of a multipotent state. Activated T cells, for instance, may interfere with dedifferentiation of monocytes by promoting their differentiation toward osteoclasts via the expression of RANKL [46].

Results from the present study and a previous publication [5] show that monocyte-derived NeoHeps are generated in a 2-step process involving dedifferentiation with passage of the monocyte through a stem cell-like progenitor, and subsequent redifferentiation. We therefore favor dedifferentiation rather than transdifferentiation as the underlying mechanism, given that both mechanisms have been recognized as separate entities [10]. Dedifferentiation of cells in adult mammals with subsequent crossing of lineage boundaries has not been clearly and unequivocally documented [10]; however, there is increasing evidence that under physiologic conditions cell dedifferentiation exists in reversible equilibrium with differentiation. For instance, in hematopoietic development postnatal hemangioblasts are generated by dedifferentiation of committed HSCs [47]. We envision that both phenotype and metabolic function of NeoHeps, and possibly other tissue-specific PCMO derivatives, can be improved by optimizing the conditions for cell-type-specific differentiation, and, even more effectively, by enhancing the PCMOs' plasticity through the above-mentioned genetic and epigenetic strategies.

Footnotes

Acknowledgments

We are indebted to J. Richter, M.Sc. (Institute of Human Genetics, UKSH, Campus Kiel) for performing the Bisulfite-Pyrosequencing analysis of OCT4 and I. Berg, M. Jansen, and K. Kopp for excellent technical assistance. We thank Drs. P. Björquist and N. Heins (CellArtis, Stockholm, Sweden) for donating RNA of the human ESC line SA002.5. This work was supported in part by a grant from the “Bundesministerium für Bildung und Forschung” (01 GN 0985).

Author Disclosure Statement

The authors declare that no competing financial interests exist.

References

Schulze

, Fändrich

, Ungefroren

, Kremer

. 2005. Adult stem cells–perspectives in treatment of metabolic diseases. Acta Gastroenterol Belg, 68:461–465.

Gordon

, Taylor

. 2005. Monocyte and macrophage heterogeneity. Nat Rev Immunol, 5:953–964.

Hutchinson

, Riquelme

, Wundt

, Hengstler

, Fändrich

, Ungefroren

, Clement

. 2007. Could treatment with neohepatocytes benefit patients with decompensated chronic liver disease? Am J Hematol, 82:947–948.

Pufe

, Petersen

, Fändrich

, Varoga

, Wruck

, Mentlein

, Helfenstein

, Hoseas

, Dressel

, Tillmann

, Ruhnke

. 2008. Programmable cells of monocytic origin (PCMO): a source of peripheral blood stem cells that generate collagen type II-producing chondrocytes. J Orthop Res, 26:304–313.

Ruhnke

, Ungefroren

, Nussler

, Martin

, Brulport

, Schormann

, Hengstler

, Klapper

, Ulrichs

, Hutchinson

, Soria

, Parwaresch

, Heeckt

, Kremer

, Fändrich

. 2005. Differentiation of in vitro-modified human peripheral blood monocytes into hepatocyte-like and pancreatic islet-like cells. Gastroenterology, 128:1774–1786.

Ruhnke

, Nussler

, Ungefroren

, Hengstler

, Kremer

, Hoeckh

, Gottwald

, Heeckt

, Fandrich

. 2005. Human monocyte-derived neohepatocytes: a promising alternative to primary human hepatocytes for autologous cell therapy. Transplantation, 79:1097–1103.

Ehnert

, Nussler

, Lehmann

, Dooley

. 2008. Blood monocyte-derived neohepatocytes as in vitro test system for drug metabolism. Drug Metab Dispos, 36:1922–1929.

Glanemann

, Gaebelein

, Nussler

, Hao

, Kronbach

, Shi

, Neuhaus

, Nussler

. 2009. Transplantation of monocyte-derived hepatocyte-like cells (NeoHeps) improves survival in a model of acute liver failure. Ann Surg, 249:149–154.

Yan

, Han

, Wang

, Liu

, Hong

, Fan

. 2007. Peripheral blood monocytes from patients with HBV related decompensated liver cirrhosis can differentiate into functional hepatocytes. Am J Hematol, 82:949–954.

10.

Wagers

, Weissman

. 2004. Plasticity of adult stem cells. Cell, 116:639–648.

11.

Grage-Griebenow

, Lorenzen

, Fetting

, Flad

, Ernst

. 1993. Phenotypical and functional characterization of Fc gamma receptor I (CD64)-negative monocytes, a minor human monocyte subpopulation with high accessory and antiviral activity. Eur J Immunol, 23:3126–3135.

12.

Contreras

, Jemionek

, Stevenson

, Hartwig

, Fauci

. 1980. An improved technique for the negative selection of large numbers of human lymphocytes and monocytes by counterflow centrifugation–elutriation. Cell Immunol, 54:215–229.

13.

Romagnani

, Annunziato

, Liotta

, Lazzeri

, Mazzinghi

, Frosali

, Cosmi

, Maggi

, Lasagni

, Scheffold

, Kruger

, Dimmeler

, Marra

, Gensini

, Maggi

, Romagnani

. 2005. CD14+CD34low cells with stem cell phenotypic and functional features are the major source of circulating endothelial progenitors. Circ Res, 97:314–322.

14.

Ungefroren

, Groth

, Fändrich

. 2006. Antioxidants and inhibitors of flavoprotein-dependent oxidases abrogate TGF-beta induction of biglycan: evidence for a role of reactive oxygen species. Signal Transduct, 6:338–344.

15.

Groth

, Schulze

, Kalthoff

, Fändrich

, Ungefroren

. 2005. Adhesion and Rac1-dependent regulation of biglycan gene expression by transforming growth factor-beta. Evidence for oxidative signaling through NADPH oxidase. J Biol Chem, 280:33190–33199.

16.

Dusi

, Donini

, Lissandrini

, Mazzi

, Bianca

, Rossi

. 2001. Mechanisms of expression of NADPH oxidase components in human cultured monocytes: role of cytokines and transcriptional regulators involved. Eur J Immunol, 31:929–938.

17.

Servet-Delprat

, Arnaud

, Jurdic

, Nataf

, Grasset

, Soulas

, Domenget

, Destaing

, Rivollier

, Perret

, Dumontel

, Hanau

, Gilmore

, Belin

, Rabourdin-Combe

, Mouchiroud

. 2002. Flt3+ macrophage precursors commit sequentially to osteoclasts, dendritic cells and microglia. BMC Immunol, 3:15.

18.

Feinberg

, Wara

, Cao

, Lebedeva

, Rosenbauer

, Iwasaki

, Hirai

, Katz

, Haspel

, Gray

, Akashi

, Segre

, Kaestner

, Tenen

, Jain

. 2007. The Kruppel-like factor KLF4 is a critical regulator of monocyte differentiation. EMBO J, 26:4138–4148.

19.

Tamura

, Kong

, Tunyaplin

, Tsujimura

, Calame

, Ozato

. 2003. ICSBP/IRF-8 inhibits mitogenic activity of p210 Bcr/Abl in differentiating myeloid progenitor cells. Blood, 102:4547–4554.

20.

Norton

, Deed

, Craggs

, Sablitzky

. 1998. Id helix-loop-helix proteins in cell growth and differentiation. Trends Cell Biol, 8:58–65.

21.

Liedtke

, Stephan

, Kögler

. 2008. Oct4 expression revisited: potential pitfalls for data misinterpretation in stem cell research. Biol Chem, 389:845–850.

22.

The International Stem Cell Initiative. 2007. Characterization of human embryonic stem cell lines by the International Stem Cell Initiative. Nat Biotechnol, 25:803–816.

23.

Takeda

, Seino

, Bell

. 1992. Human Oct3 gene family: cDNA sequences, alternative splicing, gene organization, chromosomal location, and expression at low levels in adult tissues. Nucleic Acids Res, 20:4613–4620.

24.

Kotoula

, Papamichos

, Lambropoulos

. 2008. Revisiting OCT4 expression in peripheral blood mononuclear cells. Stem Cells, 26:290–291.

25.

Bernstein

, Kamal

, Lindblad-Toh

, Bekiranov

, Bailey

, Huebert

, McMahon

, Karlsson

, Kulbokas

III , Gingeras

, Schreiber

, Lander

. 2005. Genomic maps and comparative analysis of histone modifications in human and mouse. Cell, 120:169–181.

26.

Boyer

, Plath

, Zeitlinger

, Brambrink

, Medeiros

, Lee

, Levine

, Wernig

, Tajonar

, Ray

, Bell

, Otte

, Vidal

, Gifford

, Young

, Jaenisch

. 2006. Polycomb complexes repress developmental regulators in murine embryonic stem cells. Nature, 441:349–353.

27.

Feldman

, Gerson

, Fang

, Li

, Zhang

, Shinkai

, Cedar

, Bergman

. 2006. G9a-mediated irreversible epigenetic inactivation of Oct-3/4 during early embryogenesis. Nat Cell Biol, 8:188–194.

28.

Santos-Rosa

, Schneider

, Bannister

, Sherriff

, Bernstein

, Emre

, Schreiber

, Mellor

, Kouzarides

. 2002. Active genes are tri-methylated at K4 of histone H3. Nature, 419:407–411.

29.

Cantz

, Key

, Bleidissel

, Gentile

, Han

, Brenne

, Schöler

. 2008. Absence of OCT4 expression in somatic tumor cell lines. Stem Cells, 26:692–697.

30.

Feinberg

, Cao

, Wara

, Lebedeva

, Senbanerjee

, Jain

. 2005. Kruppel-like factor 4 is a mediator of proinflammatory signaling in macrophages. J Biol Chem, 280:38247–38258.

31.

Alder

, Georgantas

III , Hildreth

, Kaplan

, Morisot

, Yu

, McDevitt

, Civin

. 2008. Kruppel-like factor 4 is essential for inflammatory monocyte differentiation in vivo. J Immunol, 180:5645–5652.

32.

Di Stefano

, Prigione

, Broccoli

. 2009. Efficient genetic reprogramming of unmodified somatic neural progenitors uncovers the essential requirement of Oct4 and Klf4. Stem Cells Dev, 18:707–716.

33.

Shimazaki

, Okazawa

, Fujii

, Ikeda

, Tamai

, McKay

, Muramatsu

, Hamada

. 1993. Hybrid cell extinction and re-expression of Oct-3 function correlates with differentiation potential. EMBO J, 12:4489–4498.

34.

Dooner

, Aliotta

, Pimentel

, Dooner

, Abedi

, Colvin

, Liu

, Weier

, Johnson

, Quesenberry

. 2008. Conversion potential of marrow cells into lung cells fluctuates with cytokine-induced cell cycle. Stem Cells Dev, 17:207–219.

35.

Charrière

, Cousin

, Arnaud

, Saillan-Barreau

, André

, Massoudi

, Dani

, Pénicaud

, Casteilla

. 2006. Macrophage characteristics of stem cells revealed by transcriptome profiling. Exp Cell Res, 312:3205–3214.

36.

Mikkelsen

, Hanna

, Zhang

, Ku

, Wernig

, Schorderet

, Bernstein

, Jaenisch

, Lander

, Meissner

. 2008. Dissecting direct reprogramming through integrative genomic analysis. Nature, 454:49–55.

37.

Zangrossi

, Marabese

, Broggini

, Giordano

, D'Erasmo

, Montelatici

, Intini

, Neri

, Pesce

, Rebulla

, Lazzari

. 2007. Oct-4 expression in adult human differentiated cells challenges its role as a pure stem cell marker. Stem Cells, 25:1675–1680.

38.

Lengner

, Welstead

, Jaenisch

. 2008. The pluripotency regulator Oct4. A role in somatic stem cells? Cell Cycle, 7:725–728

39.

Greco

, Liu

, Rameshwar

. 2007. Functional similarities among genes regulated by OCT4 in human mesenchymal and embryonic stem cells. Stem Cells, 25:3143–3154.

40.

Takahashi

, Tanabe

, Ohnuki

, Narita

, Ichisaka

, Tomoda

, Yamanaka

. 2007. Induction of pluripotent stem cells from adult human fibroblasts by defined factors. Cell, 131:861–872.

41.

Fukuchi

, Ito

, Shibata

, Kitamura

, Nakajima

. 2008. Activation of CCAAT/enhancer-binding protein alpha or PU.1 in hematopoietic stem cells leads to their reduced self-renewal and proliferation. Stem Cells, 26:3172–3181.

42.

Huangfu

, Maehr

, Guo

, Eijkelenboom

, Snitow

, Chen

, Melton

. 2008. Induction of pluripotent stem cells by defined factors is greatly improved by small-molecule compounds. Nat Biotechnol, 26:795–797.

43.

Seta

, Kuwana

. 2007. Human circulating monocytes as multipotential progenitors. Keio J Med, 56:41–47.

44.

Zhao

, Glesne

, Huberman

. 2003. A human peripheral blood monocyte-derived subset acts as pluripotent stem cells. Proc Natl Acad Sci USA, 100:2426–2431.

45.

Kuwana

, Okazaki

, Kodama

, Izumi

, Yasuoka

, Ogawa

, Kawakami

, Ikeda

. 2003. Human circulating CD14+ monocytes as a source of progenitors that exhibit mesenchymal cell differentiation. J Leukoc Biol, 74:833–845.

46.

Kotake

, Nanke

, Mogi

, Kawamoto

, Furuya

, Yago

, Kobashigawa

, Togari

, Kamatani

. 2005. IFN-gamma-producing human T cells directly induce osteoclastogenesis from human monocytes via the expression of RANKL. Eur J Immunol, 35:3353–3363.

47.

Prindull

, Fibach

. 2007. Are postnatal hemangioblasts generated by dedifferentiation from committed hematopoietic stem cells? Exp Hematol, 35:691–701.

48.

Siu

, Wong

, Chan

, Ngan

, Chan

, Cheung

. 2008. Overexpression of NANOG in gestational trophoblastic diseases: effect on apoptosis, cell invasion, and clinical outcome. Am J Pathol, 173:1165–1172.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.15 MB