Use of Ferritin Expression, Regulated by Neural Cell-Specific Promoters in Human Adipose Tissue-Derived Mesenchymal Stem Cells, to Monitor Differentiation with Magnetic Resonance Imaging In Vitro

The purpose of this study was to establish a method for monitoring the neural differentiation of stem cells using ferritin transgene expression, under the control of a neural-differentiation-inducible promoter, and magnetic resonance imaging (MRI). Human adipose tissue-derived mesenchymal stem cells (hADMSCs) were transduced with a lentivirus containing the human ferritin heavy chain 1 (FTH1) gene coupled to one of three neural cell-specific promoters: human synapsin 1 promoter (SYN1p, for neurons), human glial fibrillary acidic protein promoter (GFAPp, for astrocytes), and human myelin basic protein promoter (MBPp, for oligodendrocytes). Three groups of neural-differentiation-inducible ferritin-expressing (NDIFE) hADMSCs were established: SYN1p-FTH1, GFAPp-FTH1, and MBPp-FTH1. The proliferation rate of the NDIFE hADMSCs was evaluated using a Cell Counting Kit-8 assay. Ferritin expression was assessed with western blotting and immunofluorescent staining before and after the induction of differentiation in NDIFE hADMSCs. The intracellular iron content was measured with Prussian blue iron staining and inductively coupled plasma mass spectrometry. R2 relaxation rates were measured with MRI in vitro. The proliferation rates of control and NDIFE hADMSCs did not differ significantly (P > 0.05). SYN1p-FTH1, GFAPp-FTH1, and MBPp-FTH1 hADMSCs expressed specific markers of neurons, astrocytes, and oligodendrocytes, respectively, after neural differentiation. Neural differentiation increased ferritin expression twofold, the intracellular iron content threefold, and the R2 relaxation rate two- to threefold in NDIFE hADMSCs, resulting in notable hypointensity in T2-weighted images (P < 0.05). These results were cross-validated. Thus, a link between neural differentiation and MRI signals (R2 relaxation rate) was established in hADMSCs. The use of MRI and neural-differentiation-inducible ferritin expression is a viable method for monitoring the neural differentiation of hADMSCs.


Introduction
Most neurological disorders are caused by the loss of neurons or glial cells in the brain or spinal cord. Current therapies for these disorders are unable to replace damaged or lost neural cells. However, cell-based therapy offers the possibility of enhancing tissue repair and functional recovery in neurological disorders. Mesenchymal stem cells (MSCs) derived from bone marrow, umbilical cord blood, or adipose tissue are a promising cell source for cell-based therapies. MSCs have been used in several regenerative methods in animal models or patients with neurological diseases [1][2][3][4] and have been shown to enhance neurological recovery. Histological assays have confirmed that MSCs can differentiate along the neuronal lineage in vitro and in vivo [5][6][7]. Nevertheless, the fate of transplanted MSCs in live animals is still poorly understood. Thus, a noninvasive, real-time, sensitive, and clinically applicable method for tracking transplanted MSCs and monitoring their behavior in live animals would be useful.
Magnetic resonance imaging (MRI) is a suitable modality for the evaluation of stem cell therapy because of its excellent resolution and tissue contrast. MRI is commonly used for the noninvasive serial imaging of transplanted MSCs [8]. Many studies that have tracked transplanted MSCs in vivo have used superparamagnetic iron oxide (SPIO) particles as the MRI contrast agent [9]. However, the MRI signal hypointensity generated by these particles does not reflect the actual cell number because the iron oxide nanoparticles are diluted with each cell division. In addition, particles released from dead cells can be phagocytosed by host cells. Consequently, cell-labeling methods using SPIO particles are not suitable for the long-term monitoring of stem cell engraftment.
Genetic modification of cells in vitro to induce the expression of a reporter gene encoding an MRI-detectable probe is a novel approach to transplanted-cell imaging. The use of reporter genes for MRI-based cell tracking is advantageous for the longitudinal monitoring of cell transplants because gene expression correlates much more tightly than particle retention with cell viability and because transgene-based reporters are much less susceptible to signal loss through cell division. Ferritin is a ubiquitous intracellular protein that stores iron in a nontoxic form and releases it in a controlled manner. Ferritin overexpression for MRI visualization of transplanted cells has been assessed in several studies [10][11][12], and the results suggest that ferritin can be used to track the survival, growth, and migration of transplanted stem cells. Nonetheless, the use of ferritin overexpression to monitor the neural differentiation of transplanted stem cells noninvasively has not been investigated. In the present study, we developed an MRI imaging technique for assessing the neural differentiation of ferritin-tagged transplanted cells. Neural cell-specific promoters were used to regulate ferritin expression, and the ability to monitor cell differentiation in live cells was assessed.

Materials and Methods
Construction of neural-differentiation-inducible ferritin-expressing (NDIFE) recombinant lentiviral vectors human oligodendrocyte myelin basic protein promoter [MBPp] [14]) ( Table 1) were designed to contain the restriction enzyme sites needed for cloning. The ORFs were amplified by PCR using the templates and primers listed in S1 Table. The FTH1 ORF sequence was cloned into the pLVTHM vector (Addgene, cat. # 12247, Cambridge MA, USA) between the MluI and ClaI sites, downstream of the H1 promoter (H1p), generating the pLVTHM-H1p-FTH1 plasmid vector. The three neural-differentiationinducible promoter sequences were then cloned individually into pLVTHM-H1p-FTH1 upstream of the FTH1 sequence, replacing the H1p sequence. Three NDIFE vectors were thus generated: pLVTHM-SYN1p-FTH1, pLVTHM-GFAPp-FTH1, and pLVTHM-MBPp-FTH1 (Fig 1). The structures of the new plasmids were verified with DNA sequencing.

Cell proliferation assay
HADMSC proliferation was evaluated using the Cell Counting Kit-8 assay (cat. # C0037; Beyotime). HADMSCs were plated at a density of 1500 cells/well and cultured in 96-well plates for 24, 48, 72, or 96 h. The absorbance at 450 nm was then measured using an Epoch Multi-Volume Spectrophotometer System (BioTek, Winooski, VT, USA). All experiments were independently performed in triplicate.

Iron loading
To assess the intracellular iron content and MRI R2 relaxation rates in hADMSCs in vitro, we maintained each group of hADMSCs in the relevant medium with 200 μM ferric ammonium citrate (FAC) (cat. # 09713; Sigma-Aldrich, Shanghai, China) for 72 h before testing.

Western blot analysis
HADMSCs (3 × 10 5 ) in 6-well plates were harvested and lysed in RIPA Lysis and Extraction Buffer (cat. # 89900; Pierce, Rockford, lL, USA) containing the Halt Protease Inhibitor Cocktail (cat. # 78430; Pierce) and Halt Phosphatase Inhibitor Cocktail (cat. # 78420; Pierce). The total protein concentration in the samples was measured with a bicinchoninic acid protein assay kit (cat. # 23225; Pierce). We then added 4× loading buffer to the samples and boiled them for 5 min. Equal amounts of protein (30 μg/lane) from each group of hADMSCs were separated with 15% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto polyvinylidene difluoride membranes (Millipore). To detect human ferritin, we blocked the membranes with blocking buffer (Tris-buffered saline with 0.1% Tween-20 and 5% skim milk) at room temperature for 1 h and then incubated the membranes with a rabbit polyclonal anti-hFTH1 primary antibody (1:1000, cat. # 65080; Abcam) or a mouse monoclonal anti-β-actin antibody (1:5000, cat. # CW0096; CWBIO, Beijing, China) overnight at 4°C. The membranes were washed with TBS-T and incubated with an appropriate horseradish peroxidase-conjugated secondary antibody (1:5000, cat. # CW0102 or CW0103; CWBIO). The relative expression of ferritin in hADMSCs was analyzed semiquantitatively from the band intensity using the Quantity One software (Bio-Rad, Hercules, CA, USA), with normalization to human β-actin.

Evaluation of the intracellular iron content
The average intracellular iron content per cell was quantified using inductively coupled plasma mass spectrometry (ICP-MS; Thermo X Series II) with cell count normalization. Each group of hADMSCs from a 10-cm culture dish was washed thoroughly with PBS to remove free FAC, trypsinized, harvested, and counted (to obtain 3 × 10 6 cells/group). The cell pellets were resuspended in 50 μL of 68% (v/v) concentrated nitric acid (guaranteed reagent grade; JHD, Guangdong, China), digested at 90°C overnight, and diluted with 18-megaohm deionized water to 1000 μL.
In addition, intracellular iron accumulation and distribution in the hADMSCs were qualitatively assessed with Prussian blue iron staining. HADMSCs grown on coverslips were washed with PBS, fixed in 4% paraformaldehyde for 20 min, stained with a staining solution (2% potassium ferrocyanide and 3% HCl) for 30 min, and washed with PBS. Cell nuclei were counterstained with a nuclear fast red solution (0.1% nuclear fast red and 5% aluminum sulfate). The coverslips were mounted and examined under a microscope (BX51; Olympus).

In vitro MRI
HADMSCs (5 × 10 6 /group) were washed with PBS to remove free FAC, trypsinized, harvested, and uniformly embedded into 600 μL of 1% agarose to prepare cell phantoms in 600-μL Eppendorf tubes for in vitro MRI. T2-weighted MRI was performed on a 3T MAGNETOM Trio MRI system (Siemens, Germany), and the T2 values were determined using a multi-echo spin echo T2-weighted sequence with the following parameters: TR, 4000 ms; eight echo settings of 13, 26, 39, 52, 65, 78, 91, and 104 ms; matrix, 64 × 128; field of view, 50 × 100 mm; and a horizontal slice with 2-mm slice thickness at the center of each Eppendorf tube.

Statistical analysis
Quantitative data have been presented as the mean ± standard deviation (SD). To examine differences among the groups, one-way analysis of variance (ANOVA) was applied using the SPSS 13.0 software (SPSS Inc., Chicago, IL, USA). Differences were considered statistically significant at P < 0.05.

Overexpression of ferritin does not affect hADMSC proliferation
To test whether ferritin transgene overexpression inhibited hADMSC proliferation, cell proliferation was measured during a 96-h period (Fig 2). From 24 to 72 h, all cells underwent logarithmic growth and then entered a plateau phase from 72 h to 96 h. There were no statistically significant differences between the proliferation rates of the hADMSC groups (P > 0.05). The results suggested that ferritin overexpression did not inhibit hADMSC proliferation.

Neural differentiation increases ferritin expression in NDIFE hADMSCs
Western blotting was used to examine ferritin expression in hADMSCs. Ferritin levels in ND-NDIFE hADMSCs were~twofold higher than those in the corresponding control NDIFE hADMSCs (P < 0.05; Fig 4A). Similar results were observed with immunofluorescence staining: ferritin expression (red fluorescence) was significantly stronger in ND-NDIFE hADMSCs than in the corresponding control NDIFE hADMSCs (Fig 4B). These results indicated that neural differentiation significantly increased the activity of neural-differentiation-inducible promoters and upregulated downstream ferritin gene expression in the NDIFE hADMSCs.

Neural differentiation increases the intracellular iron concentration in NDIFE hADMSCs
The effect of ferritin expression on the intracellular iron content was assayed with ICP-MS and Prussian blue staining. In the qualitative analysis, many disperse deposits (blue granules) of accumulated intracellular iron were present in the cytoplasm of NDIFE hADMSCs. After the NDIFE hADMSCs underwent neural differentiation, large, dense deposits of accumulated iron were observed throughout the cytoplasm. The results indicated that the intracellular iron content increased dramatically relative to the iron content in undifferentiated NDIFE hADMSCs (Fig 5A).
In the ICP-MS quantitative analysis, the intracellular iron content (pg/cell) was approximately threefold higher in the ND-NDIFE hADMSCs than in the corresponding undifferentiated NDIFE hADMSCs (P < 0.05; Fig 5B). Ferritin expression appeared to increase as a result of neural differentiation. The fact that the intracellular iron content increased notably suggested that the cells' ability to internalize and store iron had improved.

Neural differentiation increases the R2 relaxation rate in NDIFE hADMSCs in vitro
T2-weighted magnetic resonance images and R2 relaxation rates (1/T2) of hADMSC agarose phantoms were acquired on a 3.0-T MAGNETOM Trio scanner. The R2 relaxation rate in ND-NDIFE hADMSCs was approximately two-to threefold higher than that in the corresponding undifferentiated NDIFE hADMSCs; the difference resulted in notable hypointensity in the T2-weighted images of ND-NDIFE hADMSCs (P < 0.05; Fig 6). These data showed that MRI could detect in vitro the differences in ferritin expression resulting from the neural differentiation of NDIFE hADMSCs.

Discussion
The objective of this study was to develop a method for monitoring neural differentiation of stem cells and lay a basis for monitoring their fates in live animals. MRI was used to measure ferritin, which was expressed under the control of a neural cell-specific promoter (neural-differentiation-inducible promoter). We hypothesized that neural differentiation would enhance the activity of the neural-differentiation-inducible promoter in NDIFE hADMSCs and that the expression of the downstream ferritin transgene would thus increase markedly, resulting in a strong increase in the intracellular iron content and the R2 relaxation rate. To test this hypothesis, we analyzed ferritin expression, the intracellular iron content, and T2-weighted images of NDIFE hADMSCs before and after their differentiation into neurons, astrocytes, or oligodendrocytes.
In this study, FTH1 overexpression did not inhibit hADMSC proliferation. These results are consistent with those of other MSC tracking studies using lentiviral ferritin transduction [11,19]. Our findings indicate that ferritin expressed from a lentiviral vector is a suitable MRI reporter for the labeling and tracking of stem cells. Neural cell-specific promoters derived from neural cell-specific marker genes (e.g., neuronal SYN1, astrocytic GFAP, and oligodendrocytic MBP) drive target gene expression nearly exclusively in specific cell types [13,14]. In this study, neural differentiation significantly enhanced ferritin expression, resulting in a notable increase in the intracellular iron concentration and R2 relaxation rate (hypointensity in the T2-weighted magnetic resonance images) in NDIFE  Differentiation increases the R2 relaxation rates of neural-differentiation-inducible ferritinexpressing human adipose tissue-derived mesenchymal stem cells. A T2-weighted image from magnetic resonance imaging (MRI) of agarose phantoms of neural-differentiation-inducible ferritinexpressing (NDIFE) and neurally differentiated (ND)-NDIFE human adipose tissue-derived mesenchymal stem cells (hADMSCs). The R2 relaxation rate of ND-NDIFE hADMSCs was significantly higher than that of the corresponding undifferentiated NDIFE hADMSCs, resulting in notable hypointensity in T2-weighted images of ND-NDIFE hADMSCs. Error bars represent the mean ± standard deviation (n = 4; *P < 0.05). hADMSCs. This finding suggests that the baseline activities of the neural cell-specific promoters in NDIFE hADMSCs increase when NDIFE hADMSCs undergo neural differentiation. Many studies have shown that MSCs have a tendency to differentiate along the neural lineage and express neural stem cell markers, such as nestin, to some extent even before induction, suggesting that MSCs are already committed to neural differentiation [20,21]. In this study, the expression of the ferritin transgene, the intracellular iron content, and the MRI R2 relaxation rate were cross-validated successfully. Our findings are consistent with the idea that an increase in the R2 relaxation rate, i.e., a shortening of the T2 relaxation time (1/R2), is directly proportional to the iron concentration [22]. Thus, a neural-differentiation-inducible promoter can serve as a "regulatory valve" that controls ferritin expression in NDIFE hADMSCs before and after neural differentiation. Thus, a link between neural differentiation and MRI signals (R2 relaxation rate) in hADMSCs was established. To the best of our knowledge, this is the first study in which MRI was used to monitor the neural differentiation of hADMSCs in vitro.
In contrast to transduced ferritin, iron oxide nanoparticles are exogenous, and their intracellular concentration cannot be regulated by neural cell-specific promoters. Thus, the correlation between the neural differentiation of hADMSCs and the MRI signals from iron oxide nanoparticles is not coincidental. On the other hand, monitoring the neural differentiation status in vivo (as opposed to in vitro) might be complicated because the neural differentiation of transplanted cells is generally almost synchronous with migration in response to a lesion in living organisms. Further in vivo research using models of nervous system diseases is warranted.
In summary, ferritin, expressed under the control of a neural-differentiation-inducible promoter, is a suitable MRI marker for monitoring the neural differentiation of hADMSCs. Furthermore, the method might have other applications. For example, inducible (specific) promoters, acting as "regulatory valves" to control the expression of target genes, could be used to evaluate the biosafety profile of stem cells and monitor the malignant transformation of tumor cells.