Induction of Expandable Adipose-Derived Mesenchymal Stem Cells from Aged Mesenchymal Stem Cells by a Synthetic Self-Replicating RNA

Adipose-derived mesenchymal stem cells (ADSCs) have attracted attention due to their potential for use in the treatment of various diseases. However, the self-renewal capacity of ADSCs is restricted and their function diminishes during passage. We previously generated induced tissue-specific stem cells from mouse pancreatic cells using a single synthetic self-replicating Venezuelan Equine Encephalitis (VEE)-reprogramming factor (RF) RNA replicon (SR-RNA) expressing the reprogramming factors POU class 5 homeobox 1 (OCT4), Krueppel-like factor 4 (KLF4), Sex determining region Y-box 2 (SOX2), and Glis Family Zinc Finger 1 (GLIS1). This vector was used to generate induced pluripotent stem (iPS) cells. Here, we applied this SR-RNA vector to generate human iTS cells from aged mesenchymal stem cells (hiTS-M cells) deficient in self-renewal that were derived from adipose tissue. These hiTS-M cells transfected with the SR-RNA vector survived for 15 passages. The hiTS-M cells expressed cell surface markers similar to those of human adipose-derived mesenchymal stem cells (hADSCs) and differentiated into fat cells and osteoblasts. Global gene expression profiling showed that hiTS-M cells were transcriptionally similar to hADSCs. These data suggest that the generation of iTS cells has important implications for the clinical application of autologous stem cell transplantation.


Introduction
Bone marrow-derived mesenchymal stem cells and adipose-derived mesenchymal stem cells (ADSCs) have attracted attention because they have the potential for use in the treatment of various diseases. Numerous published clinical studies have described their use in lower limb revascularization, liver reproduction, cartilage reproduction, and treatment of graft-versus host disease [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15]. However, the self-renewal competence of ADSCs is limited, and their function diminishes during passage [16][17][18]. In contrast, embryonic stem (ES) cells and induced pluripotent stem (iPS) cells self-renew and are pluripotent because they differentiate into cells derived from the three embryonic germ layers [19][20][21][22][23][24]. Numerous problems are associated with ES/iPS cell therapy. For example, ES cells may pose ethical issues, and ES and iPS cells may form tumors and have limited in vitro abilities to generate pure populations of differentiated cell types [25]. Therefore, we focused on developing a method for generating induced tissue-specific stem (iTS) cells derived from the pancreas (iTS-P) or liver (ITS-L) by transfecting them with a plasmid that coexpresses the reprogramming factors octamer-binding transcription factor 4 (OCT4), sex-determining region Y-box 2 (SOX2), kruppel-like factor 4 (KLF4), and c-MYC and then performing tissue-specific selection. Notably, the iTS-P/iTS-L cells were unable to generate teratomas when subcutaneously transplanted into immunodeficient mice.
Here, we applied this SR-RNA vector to generate expandable human iTS cells from aged mesenchymal stem cells (hiTS-M cells) derived from adipose tissue cells deficient in self-renewal.

Generation of hiTS-M Cells from Human ADSCs by an RNA Expression Vector
We attempted to generate hiTS-M cells from aged human ADSCs (hADSCs) that did not proliferate, cultured for 10 days after the 14th passage ( Figure 1D). We transfected the aged hADSCs with the SR-RNA encoding the reprogramming factors OCT4, KLF4, SOX2, and GLIS1 ( Figure S1). Aged hADSCs gradually proliferated until four days after transfection with the RNA vector ( Figure 1A). Nine days after transfection, the proliferating hiTS-M cells were separately passaged in a new T25 flask. The hiTS-M cells exhibited a fibroblast-like morphology similar to that of hADSCs self-renewed for 15 additional passages, for a total of 29 passages ( Figure 1A). In contrast, aged untransfected hADSCs did not proliferate ( Figure 1D). Reverse transcription-polymerase chain reaction (RT-PCR) analysis did not detect the SR-RNA in hiTS-M cells at passage five after transfection (total passages: 14 + 5 = 19) ( Figure 1B). We performed quantitative RT-PCR (qRT-PCR) analysis of marker genes of ES cells to investigate gene expression in hiTS-M cells (passage 29) and hADSCs (passage five). We were unable to detect the mRNAs encoding the pluripotency markers NANOG, OCT4, or SOX2 in either cell type ( Figure 1C). We evaluated telomerase activity in hiTS-M cells. Both hADSCs (passage 9) and hiTS-M cells (passage 14 + 9) did not express human telomerase reverse transcriptase (hTERT) ( Figure 1E). Miyagi-Shiohira.

Genes and Proteins Expressed in hiTS-M Cells
We investigated the mRNAs encoding CD73, CD105, CD55, CD59, CD71, and CD166, which are specific markers for ADSCs. hiTS-M cells (passage 14 + 6) and hADSCs (passage 6) expressed each mRNA, and the hiTS-M cells expressed significantly higher levels of CD71 mRNA. In contrast, hiTS-M cells expressed significantly lower levels of CD105 and CD55 mRNAs than hADSCs ( Figure 3A). hiTS-M cells and hADSCs expressed the mRNAs encoding insulin-like growth factor 1 (IGF1), hepatocyte growth factor (HGF), fibroblast growth factor 2 (FGF2), vascular endothelial cell growth factor A (VEGFA), and epidermal growth factor (EGF). hiTS-M cells expressed HGF and FGF2 at levels four-and six-fold higher compared with hADSCs, respectively. In contrast, hiTS-M cells expressed significantly lower levels of VEGFA and EGF mRNAs compared with hADSCs ( Figure 3B).
We also investigated expression of CD73 and CD105 protein by Flow cytometry and immunofluorescence. Both hADSCs and hiTS-M cells expressed CD73 and CD105 protein ( Figure 3C,D).
Kumar et al. showed that mesenchymal progenitors derived from human pluripotent stem cells give rise to proliferative pericytes, smooth muscle cells, and mesenchymal stem/stromal cells [9]. We evaluated which cell types hiTS-M cells included. Over 99% of hiTS-M cells did not express NG2, Calponin, or Desmin, similar to hADSCs ( Figure S2). Therefore, over 99% of hiTS-M cells were mesenchymal stem/stromal cells.

Analysis of the Differentiation Potential of hiTS-M Cells
To test whether the hiTS-M cells underwent adipogenic differentiation, the cells were treated with adipogenic induction medium for seven days and cultured in maintenance medium for an additional seven days. Oil Red O stained all hiTS-M cells ( Figure 4A), suggesting that they exhibited the adipogenic phenotype.
To determine whether the hiTS-M cells underwent osteogenic differentiation, the cells were cultured in osteogenic induction medium for 10 days.

Bisulfite Sequencing of the Genomic Promoter Regions of OCT4 and NANOG in hiPS Cells, hADSCs, and hiTS-M Cells
Bisulfite genomic sequencing demonstrated that the promoters of OCT4 and NANOG were methylated in hiTS-M cells (passage 14 + 6) and hADSCs (passage 6). In contrast, these promoters were almost completely demethylated in hiPS cells (passage 20) (Figure 5). These results demonstrate that the extents of methylation of these promoters in hiTS-M cells were similar to those of hADSCs and differed from those of hiPS cells.

Microarray Analysis
We compared the global gene expression profiles of hiPS cells (passage 20), hADSCs (passage 6), and hiTS-M cells (passage 14 + 6) (Table S3). Of 54,613 genes, 2.13% differed by more than two-fold between hADSCs and hiTS-M cells; 19.53% differed by more than two-fold between hiPS cells and hiTS-M cells; and 20.78% differed by more than two-fold between hiPS cells and hADSCs ( Figure 6A). These data suggest that the expression pattern of hiTS-M cells was similar to that of hADSCs but differed from that of hiPS cells. Unsupervised hierarchical clustering of the gene expression profiles of hiTS-M cells, hADSCs, and hiPS cells showed that hiTS-M cells clustered more closely with hADSCs than hiPS cells ( Figure 6B).

Discussion
Here, we generated hiTS-M cells using an SR-RNA vector capable of expressing the reprogramming factors OCT4, KLF4, SOX2, and GLIS1. The RNA vector was originally utilized to generate iPS cells [28]. Because hADSC culture medium was used after transfection, iPS cells were not generated in this study. Epigenetic memory is inherited from parental cells following the reprogramming of iPS cells, and the phenotypes of iPS cells may be influenced by the parental cells [30][31][32][33][34][35]. Although some of the aged hADSCs may have induced hiPS cells immediately after transfection in this study, the cells could not maintain the hiPS cell phenotype because the culture conditions were suitable for propagating hADSCs but not ES/iPS cells. Our present study shows that hiTS-M cells were generated by transient overexpression of reprogramming factors combined with hADSC culture conditions.
ADSCs have a limited self-renewal capacity, and their maximum number of passages is normally 10-15 [36,37]. We showed here that the hADSCs could be passaged 15 times after transfection (14 passages before transfection and 15 passages after transfection), and the cell number increased from 10 9 (day 0) to 10 26 (day 70) ( Figure 1D). Extending the passage number is important for autologous transplantation of ADSCs. In contrast, the characteristics of hiTS-M cells were similar but not identical to those of hADSCs as indicated by qRT-PCR and microarray analyses. The levels of mRNAs encoding CD105, CD55, VEGFA, and EGF were lower in hiTS-M cells compared with those of hADSCs, whereas those of the mRNAs encoding CD71, HGF, and FGF2 were higher in hiTS-M cells compared with those of hADSCs. This may be explained by the degree of reprogramming induced by the proteins expressed by the SR-RNA vector.
In this study, we generated human iTS-M cells using a single, synthetic, self-replicating VEE-RF RNA replicon. Other groups, including ours, have used plasmid transfection to generate iPS/iTS cells [26,27]. However, these showed a DNA integration of the plasmid of around 10-20%. RNA vector was used in this study to avoid potential integration problems.
The characteristics of ADSCs change during passage [36,37]. When aged hADSCs are strongly reprogrammed by the high expression of reprogramming factors, the characteristics of the hiTS-M cells may be similar to those of hADSCs during early passages. When aged hADSCs are weakly reprogrammed by the low expression of reprogramming factors, the characteristics of the hiTS-M cells may be similar to those of late passage hADSCs. The characteristics of hiTS-M cells may depend on the degree of reprogramming and their passage number after transfection. As shown in Figure 4, both hiTS-M and hADSCs appear to have partially lost the ability to differentiate. This may be due to the variability in hADSCs and in the reprogramming of hiTS-M cells.
In conclusion, we generated hiTS-M cells from aged hADSCs. hiTS-M cells expressed cell surface markers similar to those of hADSCs and were induced to differentiate into fat cells and osteoblasts. qRT-PCR and bisulfite genomic sequencing analyses of OCT4 and NANOG showed that the phenotype of hiTS-M cells differed significantly from that of hiPS cells. Global gene-expression profiling showed that hiTS-M cells are transcriptionally similar to hADSCs. Because the self-renewal capacity of hADSCs is restricted and the function of hADSCs decreases during passage, the generation of hiTS-M cells may have important implications for the clinical application of autologous stem cell transplantation.

Culture of hADSCs
Human ADSCs derived from a 40-year-old female donor (Lot No. 0000353102; body mass index, 23; passage 1) were obtained from Lonza Japan, Inc. (Tokyo, Japan). For adherent cell culture, 4 × 10 5 hADSCs were seeded into a T-25 flask (Asone, Osaka, Japan) using the Poietics ADSC-GM BulletKit (PT-4505; Lonza Japan) at 37 • C in a humidified atmosphere containing 5% CO 2 . The cells were cultured until they were approximately 80-90% confluent. The cells were detached from the culture flask using trypsin-ethylenediaminetetraacetic acid (EDTA) (0.25% w/v; Wako, Osaka, Japan) and repeatedly seeded and cultured in new culture flasks. The hADSCs were passaged 14 times until they failed to proliferate. The hADSCs, from passages four to seven, were used as a control for the analyses as follows: RT-PCR, qRT-PCR, differentiation into adipocytes and osteocytes, bisulfite genomic sequencing, and microarray analysis.

Generation of hiTS-M Cells Using an RNA Replicon
The hiTS-M cells were generated as described previously [28], with minor modification, using a Simplicom RNA Reprogramming Kit (Merck Millipore, Tokyo, Japan). Aged hADSCs (passage 14) were seeded into a T25 plate on day-10 (40-50% confluency). To minimize the interferon response, cells were treated with 1 mL of Advanced DMEM (Thermo Fisher Scientific, Waltham, MA, USA) containing 0.2 µg of B18R protein (Merck Millipore, Burlington, MA, USA) 2 h before transfection. On day 0, 0.5 µg VEE-OKS-iG and 0.5 µg B18R mRNAs was used to transfect cells in the presence of Lipofectamine 2000 (Thermo Fisher Scientific). After 3 h, the transfection medium was changed to Advanced DMEM containing 200 ng/mL of B18R protein. The cells were passaged on day 9, and Advanced DMEM containing 200 ng/mL of B18R protein was supplied every day until day 16. Cells were then passaged on day 16 and Advanced DMEM was replaced with hADSC culture medium ( Figure S1).

RT-PCR and qRT-PCR Analyses
Total RNA was extracted from cells using an RNeasy Mini Kit (Qiagen, Tokyo, Japan). After quantifying the RNA using spectrophotometry, 2.5 µg RNA was heated at 85 • C for 3 min and then reverse-transcribed in a 25-µL solution containing 200 units of Superscript II RNase H-RT (Thermo Fisher Scientific), 50 ng random hexamers (Thermo Fisher Scientific), 160 µmol/L dNTP, and 10 nmol/L dithiothreitol. The reaction conditions were as follows: 10 min at 25 • C, 60 min at 42 • C, and 10 min at 95 • C. Amplifications were performed using a Perkin-Elmer 9700 Thermocycler (Perkin-Elmer, Waltham, CT, USA) with 3 µL of cDNA (20 ng DNA equivalents), 160 µmol/L of cold dNTPs, 10 pmol of appropriate oligonucleotide primers, 1.5 mmol/L of MgCl 2 , and 5 units of AmpliTaq Gold DNA polymerase (Perkin-Elmer, Norwalk, CT, USA) in 1× PCR buffer. The thermal cycle profile was as follows: 10 min denaturation at 94 • C followed by an amplification cycle (1 min denaturation at 94 • C, 1 min annealing at 57 • C to 62 • C, and a 1 min extension at 72 • C) with a final extension step of 10 min at 72 • C. The oligonucleotide primers are shown in Table S1.
The mRNAs were quantified using a TaqMan Real-time PCR System (Applied Biosystems, Foster City, CA, USA) according to the manufacturer's instructions. PCR was performed for 40 cycles, including 2 min at 50 • C and 10 min at 95 • C as initial steps. In each cycle, denaturation was performed for 15 s at 95 • C, and annealing and extension were performed for 1 min each at 60 • C. PCR was performed in 20 µL of a solution of cDNAs synthesized from 1.11 ng of total RNA. The level of each mRNA sample was normalized by dividing its amount by that of GAPDH mRNA. Primers for the mRNAs encoding OCT4, SOX2, NANOG, TERT, 5 -nucleotidase (CD73), endoglin (CD105), complement decay accelerating factor (CD55), complement protectin (CD59), transferrin receptor (CD71), and activated lymphocyte cell adhesion molecule (CD166); primers specific for the genes encoding IGF1, HGF, FGF2, VEGFA, EGF, RUNX2, and GAPDH were included in Assays-on-Demand Gene Expression Products (Thermo Fisher Scientific).

Immunostaining
Cell were fixed with 4% paraformaldehyde in PBS buffer. After blocking with 20% AquaBlock (EastCoast Bio, Inc. North Berwick, ME, USA) for 30 min at room temperature, the cells were incubated overnight at 4 • C with each 1st antibody (FITC anti-human CD73 (Ecto-5 -nucleotidase) Antibody, Alexa Fluor 647 anti-human CD105 Antibody, Alexa Fluor 488 anti-human NG2 Antibody, mouse anti-human Calponin Antibody, and rabbit anti-Desmin Antibody) and then for 1 h at room temperature with each second antibody (AlexaFluor 488 anti-mouse antibody, AlexaFluor 488 anti-rabbit antibody). Mounting medium for fluorescence with DAPI (Vector Laboratories Inc., Burlingame, CA, USA) was used for mounting.

Cell Differentiation
Adipogenic differentiation was performed using Adipogenic Differentiation Medium (DM-2; Zen-Bio, Inc., Research Triangle Park, NC, USA) and Adipocyte Maintenance Medium (AM-1; Zen-Bio, Inc.). Confluent hiTS-M cells (passage 14 + 7) and hADSCs (passage 7) (three to four days after seeding into six-well plates using the designated medium) were cultured for seven days using Adipocyte Differentiation Medium (DM-2; Zen-Bio, Inc.) to induce differentiation into adipocytes. The medium was subsequently switched to Adipocyte Maintenance Medium (AM-1; Zen-Bio, Inc.) and changed every three days. Adipocytes with lipid droplets were detected using light microscopy after 20-30 days. Differentiation was confirmed by Oil Red O (Cosmo Bio Co., Ltd., Tokyo, Japan) staining of intracellular lipid droplets. Differentiated ASCs were fixed in 10% formaldehyde (Wako, Osaka, Japan) in PBS (Wako, Osaka, Japan) for at least 10 min, washed with 60% isopropanol (Wako, Osaka, Japan), and stained with Oil Red O (Wako, Osaka, Japan) for 10 min followed by repeated washing with water and destaining in 100% isopropanol for 1 min [14,25,39].
Osteogenic differentiation was induced by culturing the hiTS-M cells and hADSCs (passage 3) for 10 days in MSC Go Rapid Osteogenic XF TM (Cosmo Bio Co., Ltd., Tokyo, Japan). Differentiation was determined through detection of extracellular matrix calcification as revealed by staining with Alizarin Red solution (ARD-A1; PG Research Co., Ltd., Tokyo, Japan) according to the manufacturer's instructions [10,14,25,39,40].

Bisulfite Genomic Sequencing
Bisulfite treatment was performed using the CpGenome Turbo Bisulfite Modification Kit (Merck Millipore) according to the manufacturer's recommendations. The PCR primers are listed in Table S2. Amplicons were cloned using a Mighty TA-cloning kit (TAKARA BIO INC., Shiga, Japan). Ten randomlyselected clones were sequenced using M13 forward and reverse primers specific for each gene.

Microarray Analysis
Total RNA from ES cells, iTS-P cells, or islets was labeled with biotin. Samples were hybridized to the GeneChip 3 IVT PLUS Reagent Kit (Affymetrix, Tokyo, Japan) and GeneChip Hybridization, Wash and Stain Kit (Affymetrix, Tokyo, Japan) according to the manufacturer's protocol. Arrays were scanned using a GeneChip Scanner 3000 7G (Affymetrix, Tokyo, Japan). Data were analyzed using Transcriptome Analysis Console 4.0 software (Thermo Fisher).

Statistical Analyses
The data are expressed as the mean ± standard error. Two groups were compared using the Student t-test. The differences between each group were considered significant if p < 0.01. All statistical analysis methods were performed in accordance with the relevant guidelines and regulations.