Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print August 23, 2007, 10.1148/radiol.2443061290
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2443061290v1
245/1/206    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ju, S.
Right arrow Articles by Ni, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ju, S.
Right arrow Articles by Ni, Y.
(Radiology 2007;245:206-215.)
© RSNA, 2007

Molecular Imaging

In Vivo MR Tracking of Mesenchymal Stem Cells in Rat Liver after Intrasplenic Transplantation1

Shenghong Ju, MD, PhD, Gao-Jun Teng, MD, Haihua Lu, MD, Yu Zhang, PhD, Aifeng Zhang, PhD, Feng Chen, MD, and Yicheng Ni, MD, PhD

1 From the Laboratory of Molecular Imaging, Department of Radiology, Zhongda Hospital (S.J., G.J.T., H.L., F.C.), Laboratory of Molecular and Biomolecular Electronics (Y.Z.), and School of Basic Medical Science (A.Z.), Southeast University, 87 Ding Jia Qiao Road, Nanjing 210009, China; and Department of Radiology, University Hospitals, Catholic University of Leuven, Leuven, Belgium (F.C., Y.N.). Received July 26, 2006; revision requested September 27; revision received November 20; accepted December 20; final version accepted February 12, 2007. Supported by National Nature Science Foundation of China grant 30400116 and Natural Science Foundation of Jiangsu Province grant Bk2004068. Address correspondence to G.J.T. (e-mail: gjteng{at}vip.sina.com).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Purpose: To prospectively track in vivo in rats intrasplenically transplanted stem cells labeled with superparamagnetic particles by using magnetic resonance (MR) imaging.

Materials and Methods: The study was approved by the institutional Committee on Animal Research. Liver damage in 12 rats was induced with subcutaneous injection of carbon tetrachloride (CCl4). Intrasplenic transplantation of 6 x 106 rodent bone mesenchymal stem cells (BMSCs) with (n = 6) and without (n = 6) superparamagnetic particle Fe2O3-poly-L-lysine (PLL) labeling was performed via direct puncture. Cell labeling efficiency was assessed in vitro by using Prussian blue stain and an atomic absorption spectrometer. MR examinations were performed immediately before and 3 hours and 3, 7, and 14 days after transplantation. Liver-to-muscle contrast-to-noise ratios (CNRs) on T2*-weighted MR images obtained before and after injection were measured and correlated with histomorphologic studies. Statistical analyses were performed by using repeated-measures analysis of variance.

Results: Rat BMSCs could be effectively labeled with approximately 100% efficiency. Migration of transplanted labeled cells to the liver was successfully documented with in vivo MR imaging. CNRs on T2*-weighted images decreased significantly in the liver 3 hours after injection of BMSCs (P < .05) and returned gradually to the level achieved without labeled cell injection in 14 days. Histologic analyses confirmed the presence of BMSCs in the liver. The labeled cells primarily localized in the sinusoids of periportal areas and the foci of CCl4-induced liver damage. Quantitative analysis of Prussian blue-stained cells indicated gradual decrease of dye pigments from 3 hours to 3, 7, and 14 days after injection. No free iron particles were found in the interstitium or within hepatic microvessels.

Conclusion: The rat BMSCs could be efficiently labeled with Fe2O3-PLL and the relocation of the labeled cells to rat livers after intrasplenic transplantation could be depicted at in vivo MR imaging.

© RSNA, 2007


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
As a potential interventional procedure, liver cell transplantation provides an effective strategy for the treatment of an impaired liver or liver failure (1). When compared with orthotopic liver transplantation, cell transplantation has the advantages of lower cost, lower risk, and simpler manipulation of the procedure. Autologous cell transplantation helps prevent immunologic rejection, which is always a problem for orthotopic liver transplantation, and it bears great potential usefulness as a transient therapy before liver transplantation (1).

However, there remain many important issues to be addressed regarding cell therapy. The first issue is the shortage of resources in mature liver cells. Exploitation of high regeneration and differentiation of stem cells may solve this problem. Transdifferentiation of different kinds of stem cells into hepatocytes has been previously demonstrated with embryonic stem cells, hepatoblasts (mostly in fetal liver), hepatic oval cells (mostly in mature liver), pancreatic progenitor cells, bone marrow hematopoietic stem cells, and mesenchymal stem cells (MSCs) (2–5). Bone marrow–derived stem cells exhibited better prospects in terms of practical applicability as surrogates of hepatocytes for transplantation in MSCs. Both in vitro and in vivo studies appeared to support bone MSCs (BMSCs) as being more potent in hepatocytic transdifferentiation (6–9).

The second issue in liver cell transplantation is monitoring migration, distribution, and evolution of the transplanted cells. Since conventional histologic examinations such as light microscopy with hematoxylin-eosin staining are unable to help distinguish between transplanted donor cells and the recipient cells, stem cells are usually tagged in vitro and then transplanted in vivo. The tissue or organ has to be removed at certain time points and processed with special histochemical or immunochemical procedures to enable researchers to visualize the tagged cells (10).

In 1992, Soriano et al (11) marked hepatocytes with fluorescent dye DiI (Molecular Probes, Eugene, Ore) and successfully identified the transplanted hepatocytes by using fluorescent microscopy. However, those tagging methods require in vitro preparations and examinations of histologic materials, which are unsuitable for noninvasive and repeated monitoring of in vivo transplanted stem cells in humans. Therefore, more recent research activities have focused on in vivo real-time tracking and detecting the fate of transplanted stem cells by using appropriate imaging technologies (10,12–14). Thus, radiologists may play an increasingly crucial role in future experimentation and application.

It is well known that most hepatocytes migrate and repopulate in the liver after intrasplenic transplantation (15–17). However, the process of migration and repopulation of the stem cells from the spleen to the liver has not been tracked in vivo, to our knowledge. Thus, the purpose of our study was to prospectively track in rats in vivo intrasplenic transplantation of stem cells labeled with superparamagnetic particles during magnetic resonance (MR) imaging.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Synthesis of Ferric Oxide Nanoparticles and Coupling with Poly-L-Lysine
A mixture of Fe2+ and Fe3+ ions with pH of 1.7 was prepared under nitrogen gas protection and was titrated to pH of 9 with 1.5 mol/L ammonia and vigorous stirring. After a reaction time of 30 minutes, black Fe3O4 precipitate was obtained. The precipitate was separated by using a permanent magnet, washed repeatedly with deionized water, and then diluted into a water-based magnetic fluid containing a concentration of Fe3O4 at 3 g/L. Fe(NO3)3 was used to oxidize the Fe3O4 magnetic fluid in an acidic condition at 90°C to obtain Fe2O3 magnetic fluid, which was prepared by using the same magnetic separation method, and was diluted to 3 g/L with deionized water (18,19). An aqueous solution of 1% poly-L-lysine (PLL) was added to the Fe2O3 magnetic fluid and the reaction was sustained for 2 hours by stirring to obtain PLL-wrapped Fe2O3 nanoparticles. The product was adjusted to a 3 g/L Fe2O3-PLL water-based magnetic fluid. The amount of PLL that adhered to the particle surface was 0.01% of total Fe2O3 mass.

Cell Culture
Our study was approved by the institutional Committee on Animal Research. The BMSCs were acquired from rat bone marrow. After six Sprague-Dawley rats were euthanized with an overdose of pentobarbital, the femurs and tibias were crushed and rinsed with a Dulbecco's modified Eagle's medium (Gibco, Karlsruhe, Germany) to harvest resuspended bone marrow cells. The cells were placed in culture dishes of low-glucose Dulbecco's modified Eagle's medium supplemented with 0.2 mmol/mL L-glutamine (Gibco, Carlsbad, Calif), 100 U penicillin (Nanjing Jinling Pharmaceutical, Nanjing, China), 100 µg/mL streptomycin (Nanjing Jinling Pharmaceutical), and 10% fetal bovine serum (Worthington Biochemical, Lakewood, NJ). This mixture was adjusted to 106 cells per milliliter of the concentration, placed in a 25-cm2 culture flask, and grown in a standard culture medium at 37°C with 5% CO2. Three days later, the culture solution was replaced for the first time, and the cells that did not adhere to the flask wall were removed.

The cultured solution was replaced every 3 days for an average of three times until the number of cells adhering to the flask wall reached approximately 90% of the flask wall area. The original medium was discarded, and 1.5 mL of 0.25% trypsin was added to the flask to separate the cells from the flask bottom. After fresh culture medium was added, the cells were passed in a 1:2 ratio and marked as passage 1. During the passage, the culture medium was replaced every 3 days until the cells were grown to confluence. The procedure was repeated and the cells were marked as passage 2, passage 3, and so on. The BMSCs were thus selected (J.S.) on the basis of adherence to plastics as one of the physicochemical characteristics of these cells (20).

The membranous antigen expressed on BMSCs was determined by using fluorescence-activated cell sorter analysis with a flow cytometer (FACS Caliber; Becton Dickinson, San Jose, Calif) equipped with Cell Quest software (Becton Dickinson). Cells were harvested and stained (J.S., Z.A.) for cell surface antigens CD45, CD90, and CD29 by using phycoerythrin or fluorescein isothiocyanate–marked mouse antirat CD45, CD90, and CD29 antibodies (Caltag Laboratories, Burlingame, Calif; eBioscience, Biolegend, San Diego, Calif) or an isotype-matched control.

Multipotency was assessed by measuring osteogenic properties, neural assays, and hepatocyte-like differentiation. To measure osteogenic properties, an alkaline phosphatase detection kit was used to detect enzyme activity for both passage 6 and passage 1 BMSCs (21). To measure neural assays (22,23), an induction medium containing 20 ng/mL basic fibroblast growth factor, 10 ng/mL epidermal growth factor, and 2% B27 medium supplement (Gibco, Karlsruhe, Germany) was added to the standard medium for 7 days. After formalin fixation, immunofluorescent staining was performed to detect the expression of cellular nestin activity by using rabbit monoclonal antibodies (Zhongshan Bioengineering, Beijing, China). To induce hepatocyte-like differentiation (6), an induction medium containing 25 ng/mL hepatocyte growth factor (Sigma-Aldrich, St. Louis, Mo) was added to the standard medium for 21 days. Immunocytochemical staining was performed (J.S., Z.A.) to detect the expression of cellular {alpha}-fetoprotein and albumin activity with goat antirat {alpha}-fetoprotein (Genetimes Bioengineering, Shanghai, China) and albumin antibodies (Santa Cruz Biotechnology, Santa Cruz, Calif).

Cell Labeling, Prussian Blue Staining, and Spectrometry
A culture medium containing Fe2O3-PLL was added to the BMSCs of passage 3 for labeling in 25-cm2 flasks. The final iron concentration of 20 µg/mL was chosen for labeling BMSCs on the basis of previous tests (14). The cell cultures were kept for 12 hours at 37°C in a 95% air per 5% CO2 incubator.

After incubation, the BMSCs were washed with phosphate-buffered saline three times to remove excess Fe2O3-PLL. For Prussian blue staining, the cells were incubated with 2% potassium ferrocyanide in 6% hydrochloric acid (HCl) and then counterstained with nuclear fast red, revealing the cells' outlines and intracellular iron particles. To measure iron concentration within the cells, cell precipitates were dissolved in 37% HCl and analyzed (Z.A.) by using a polarized atomic absorption spectrometer (HG-9602A; Shengyang Huaguang, Shengyang, China). The remaining BMSCs were used for animal transplantation experiments.

Rat Models with Persistent Liver Damage
To generate liver damage, 1 mL/kg carbon tetrachloride (CCl4) in olive oil (at a 1:1 ratio) was administered with subcutaneous hypodermic injection twice a week to 12 rats at dorsum over the course of 4 weeks (L.H., T.G.) to induce persistent liver damage and subsequent cirrhosis (24).

Intrasplenic Transplantation of MSCs
One day after the fourth administration of CCl4, the abdomens of the cirrhotic rats were opened to expose the spleen. Approximately 6 x 106 BMSCs suspended in 0.5 mL phosphate-buffered saline were slowly injected (L.H., T.G.) into the inferior tip of the spleen in group 1 (Fe2O3-PLL–labeled BMSCs, n = 6) and group 2 (unlabeled BMSCs, n = 6). A 30-gauge needle was used for the procedure. The pinhole at the injection site was pressed for hemostasis. One day before the operation and every other day after the operation, cyclosporine A was administered with intraperitoneal injection at a dose of 10 mg/kg to inhibit orthotopic rejection.

MR Imaging and Data Acquisition
The rats underwent MR imaging of the liver immediately before, 3 hours after, and 3, 7, and 14 days after injection of cells. MR imaging was performed with a 1.5-T imaging device (Eclipse; Philips Medical Systems, Best, the Netherlands) with a 12.7-cm receiver surface coil. The rats were anesthetized by using intraperitoneal injection of pentobarbital sodium at 30 mg/kg, were placed supine in a plastic holder, and were wrapped in towels to maintain body temperature.

Coronal T1-weighted spin-echo sequence was used as a localizer. Transverse T1-weighted spin-echo (repetition time msec/echo time msec, 500/17.9; section thickness, 2 mm; no intersection gap; matrix, 192 x 256; field of view, 80 x 80 mm), T2-weighted fast spin-echo (3000/108; section thickness, 2 mm; no intersection gap; matrix, 192 x 256; field of view, 80 x 80 mm), and T2*-weighted gradient-echo (620/15.7; section thickness, 2 mm; no intersection gap; matrix, 192 x 192; field of view, 80 x 80 mm; flip angle, 35°) sequences were performed.

A coronal T2*-weighted gradient-echo sequence was also performed by using the same parameters as the transverse T2*-weighted gradient-echo sequence. The time of acquisition (two signals acquired) was 3 minutes 12 seconds for the T1-weighted sequence, 3 minutes 30 seconds (three signals acquired) for the T2-weighted spin-echo sequence, and 3 minutes 29 seconds (three signals acquired) for the T2*-weighted sequence. All transverse sequences were performed by using the same geometry for location, field of view, and number of sections to maintain comparability between the different imaging sequences.

The average signal intensities (SIs) of the liver and the dorsal muscle before and after injection of cells were measured by one investigator (L.H.) who was blinded to the applied labeling procedure. To minimize measurement bias, the total area of the liver and dorsal muscle was covered with contoured regions of interest on each section. Liver-to-muscle contrast-to-noise ratio (CNR) was calculated as follows: (SIL – SIM)/SIN, where SIL = SI of liver, SIM = SI of muscle, and SIN = SI of background noise, which was measured anterior to the depicted liver (25). The CNRs were averaged over all liver-related sections for different time points.

Preparation for Tissue Microscopy
At the time points of 3 hours, 3 days, and 7 days after MR imaging, one rat from each group was sacrificed for histologic tissue examination. At 14 days, the rest of the rats were euthanized after MR examination. The livers were perfused via the heart with 4% paraformaldehyde for fixation, embedded in paraffin, and sectioned into 5–10-µm-thick slices for hematoxylin-eosin staining to display tissue morphology and for Prussian blue staining to identify iron particles (6).

Determination of BMSC Count in the Liver
To determine the number of accumulated BMSCs in the liver, representative Prussian blue-stained liver slides from groups 1 and 2 were examined at each time point. At least five fields of view with original magnification (x40) were randomly selected and photomicrographed. Prussian blue positively stained cells were manually counted in each view by one blinded investigator (A.Z.).

Statistical Analysis
Statistical analyses were performed with the SAS software (version 8.0; SAS Institute, Cary, NC). Numeric data including SI, CNR, and BMSC count were reported as means ± standard deviation. For statistical comparisons, repeated-measures analysis of variance was used. A P value of less than .05 was considered to indicate a significant difference.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Fe2O3-PLL Nanometer Particles
The magnetic particles obtained were almost spherical, with iron core size distribution measuring 15 nm ± 5 (Fig 1). The saturation magnetization was 60 emu/g, lower than the saturization magnetization of the corresponding bulk of magnetic particles owing to their small sizes in the nanometer dimensions (19). Cytotoxicity and oxidative effects on Chinese Hamster lung fibroblast cells in vitro and the toxicity and in vivo biodistribution of the nanoparticles have previously been reported (26–28). The Fe2O3-PLL magnetic fluids obtained were stable for 3 months when stored at 4°C.


Figure 1
View larger version (141K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1: Transmission electron microscopic image (magnification, x105) and electron diffraction pattern (inset) shows Fe2O3 nanoparticles.

 
BMSC Characterization
Twenty-four hours after transplantation, BMSCs could be seen in newly formed colonies, which were mixed with osteoblast-like cells. The BMSCs were rapidly growing fibroblast-like cells with a single nucleus and long, slim, processes at two ends with high refraction and radiating arrays. After the first passage, they looked like spindles or asters with a slim body. At passage 3, most of the miscellaneous cells were eliminated, and the remaining uniform fibroblast-like cells were BMSCs (Fig 2).


Figure 2
View larger version (116K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2: Inverse microscopic image (magnification, x200) shows BSMCs in passage 3 appearing as uniform fibroblast-like cells with slim bodies.

 
Flourescence-activated cell sorting revealed that the majority of cells expressed the MSC surface markers CD90 (99.84%) and CD29 (99.88%), but did not express the hematopoietic origin marker of CD45 (99.87%) (Fig 3). The cells could also be differentiated into alkaline phosphatase-positive osteoblasts or nestin-positive neural cells in the induction medium (Figs 4, 5). Immunocytochemical staining also indicated the expression of {alpha}-fetoprotein and albumin at 21 days (Fig 6).


Figure 3A
View larger version (26K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3a: (a–c) Graphs of fluorescence-activated cell sorter analysis show majority of cells (over 99%) express markers of (b) CD29 and (c) CD90, but not (a) CD45.

 

Figure 3B
View larger version (26K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3b: (a–c) Graphs of fluorescence-activated cell sorter analysis show majority of cells (over 99%) express markers of (b) CD29 and (c) CD90, but not (a) CD45.

 

Figure 3C
View larger version (25K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3c: (a–c) Graphs of fluorescence-activated cell sorter analysis show majority of cells (over 99%) express markers of (b) CD29 and (c) CD90, but not (a) CD45.

 

Figure 4A
View larger version (121K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4a: (a, b) Photomicrographs show alkaline phosphatase enzyme activity as black particles in cytoplasm by using detection kit. Activity (a) present for passage 6 BSMCs, but (b) absent for passage 1 BSMCs (magnification, x400).

 

Figure 4B
View larger version (107K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4b: (a, b) Photomicrographs show alkaline phosphatase enzyme activity as black particles in cytoplasm by using detection kit. Activity (a) present for passage 6 BSMCs, but (b) absent for passage 1 BSMCs (magnification, x400).

 

Figure 5
View larger version (9K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5: Microscopic image of immunofluorescent red staining after 7 days induction shows positive expression of cellular nestin activity (magnification, x100).

 

Figure 6A
View larger version (110K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6a: Hepatocyte-like differentiation. Photomicrographs show expressions of cellular (a) {alpha}-fetoprotein and (b) albumin activity at immunocytochemical staining (brown particles) in cytoplasm after 21 days induction with hepatocyte growth factor (magnification, x400).

 

Figure 6B
View larger version (93K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6b: Hepatocyte-like differentiation. Photomicrographs show expressions of cellular (a) {alpha}-fetoprotein and (b) albumin activity at immunocytochemical staining (brown particles) in cytoplasm after 21 days induction with hepatocyte growth factor (magnification, x400).

 
Prussian Blue Staining and Iron Quantification in Cell Culture
BMSCs stained with Prussian blue showed results in all labeled cells; the labeled rate was approximately 95% (Fig 7). No blue particles were seen in the unlabeled cells. The per-cell iron content of 5.46 pg ± 0.98 was determined by using an atomic absorption spectrometer.


Figure 7
View larger version (121K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7: Photomicrograph of passage 3 BMSCs incubated with Fe2O3-PLL shows intracytoplasmic particles (blue). The labeling rate reaches almost 95%. (Prussian blue stain; original magnification, x200.)

 
MR Imaging and Analysis
Liver SI on T2*-weighted images sharply decreased 3 hours after intrasplenic transplantation of labeled BMSCs, but gradually approached close to normal level at day 14. The SI decline was significantly stronger on T2*-weighted images than on T2-weighted images (P < .05) (Fig 8).


Figure 8A
View larger version (121K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 8a: T2-weighted gradient-echo MR images show SI changes of liver after injecting labeled (top row) and unlabeled (bottom row) BMSCs (a) 3 hours before and (b) 3 hours and (c) 3, (d) 7, and (e) 14 days after injection. Note gradual return of SI compared with SI before transplantation and in control rats.

 

Figure 8B
View larger version (111K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 8b: T2-weighted gradient-echo MR images show SI changes of liver after injecting labeled (top row) and unlabeled (bottom row) BMSCs (a) 3 hours before and (b) 3 hours and (c) 3, (d) 7, and (e) 14 days after injection. Note gradual return of SI compared with SI before transplantation and in control rats.

 

Figure 8C
View larger version (112K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 8c: T2-weighted gradient-echo MR images show SI changes of liver after injecting labeled (top row) and unlabeled (bottom row) BMSCs (a) 3 hours before and (b) 3 hours and (c) 3, (d) 7, and (e) 14 days after injection. Note gradual return of SI compared with SI before transplantation and in control rats.

 

Figure 8D
View larger version (122K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 8d: T2-weighted gradient-echo MR images show SI changes of liver after injecting labeled (top row) and unlabeled (bottom row) BMSCs (a) 3 hours before and (b) 3 hours and (c) 3, (d) 7, and (e) 14 days after injection. Note gradual return of SI compared with SI before transplantation and in control rats.

 

Figure 8E
View larger version (138K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 8e: T2-weighted gradient-echo MR images show SI changes of liver after injecting labeled (top row) and unlabeled (bottom row) BMSCs (a) 3 hours before and (b) 3 hours and (c) 3, (d) 7, and (e) 14 days after injection. Note gradual return of SI compared with SI before transplantation and in control rats.

 
Before injection and 3 hours and 3, 7, and 14 days after injection, the respective liver-to-muscle CNRs were –2.62 ± 2.78, –18.87 ± 1.88, –14.81 ± 1.42, –11.90 ± 2.95, and –6.65 ± 4.48 for group 1 and –0.62 ± 2.63, –1.95 ± 1.75, –2.68 ± 0.68, –1.21 ± 2.00, and –2.75 ± 3.90 for group 2. Injection of labeled cells caused a significant CNR decline in the liver for group 1 at both 3 hours and 3 days after injection compared with group 1 rats before transplantation or group 2 control rats (P < .05) (Fig 9). The CNR was still low at both 7 and 14 days, but was not significant (P > .05). For the control group, there was no statistically significant difference in CNR at each time point after transplantation compared with before (P > .05).


Figure 9
View larger version (11K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 9: Graph compares liver-to-muscle CNRs between groups 1 and 2. Injection of labeled cells caused substantial decline of CNRs at 3 hours and 3 days after injection compared with CNR before transplantation and control rats (P< .05). CNR still low at 7 and 14 days, respectively, but insignificant (P> .05).

 
Histologic Observation
Liver injury and cirrhosis were proved by using histologic review in all rats treated with CCl4. At light microscopy with hematoxylin-eosin staining, typical histopathologic changes indicative of persistent liver damage were observed (Fig 10). Coagulative necrosis of hepatocytes occurred primarily in centrilobular zones with light dot necrosis diffusely distributed to midlobular and periportal areas. Hydropic degeneration was marked by clear cytoplasm or vacuolization (ballooning). In cells with steatosis, the cytoplasm was occupied by circular vacuoles varying from a few droplets of variable size to diffuse involvement with the nuclei moved from their normal position in the cell. The injury also spread toward the intermediate zone. Several cells appeared shrunken and exhibited oxyphilous change with deeply stained cytoplasts suggestive of apoptosis. Inflammatory infiltration with mainly macrophages and lymphocytes could also be seen in areas with hepatic injury. Liver cirrhosis was evidenced by fibrotic septa formation starting in the portal areas.


Figure 10A
View larger version (166K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 10a: Photomicrographs show obvious liver injury in rats. After six hypodermic injections in 3 weeks, (a, b) main lesions include hepatic cell steatosis and hydropic change in center of hepatic lobule. (Hematoxylin-eosin stain; magnification in a, x100; magnification in b, x200.) (c, d) Shuttle-like cells positively identified 7 days after cell transplantation in cytoplast particles (blue) and nucleus (red). (Prussian blue stain; magnification in c, x100; magnification in d, x200.)

 

Figure 10B
View larger version (162K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 10b: Photomicrographs show obvious liver injury in rats. After six hypodermic injections in 3 weeks, (a, b) main lesions include hepatic cell steatosis and hydropic change in center of hepatic lobule. (Hematoxylin-eosin stain; magnification in a, x100; magnification in b, x200.) (c, d) Shuttle-like cells positively identified 7 days after cell transplantation in cytoplast particles (blue) and nucleus (red). (Prussian blue stain; magnification in c, x100; magnification in d, x200.)

 

Figure 10C
View larger version (166K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 10c: Photomicrographs show obvious liver injury in rats. After six hypodermic injections in 3 weeks, (a, b) main lesions include hepatic cell steatosis and hydropic change in center of hepatic lobule. (Hematoxylin-eosin stain; magnification in a, x100; magnification in b, x200.) (c, d) Shuttle-like cells positively identified 7 days after cell transplantation in cytoplast particles (blue) and nucleus (red). (Prussian blue stain; magnification in c, x100; magnification in d, x200.)

 

Figure 10D
View larger version (133K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 10d: Photomicrographs show obvious liver injury in rats. After six hypodermic injections in 3 weeks, (a, b) main lesions include hepatic cell steatosis and hydropic change in center of hepatic lobule. (Hematoxylin-eosin stain; magnification in a, x100; magnification in b, x200.) (c, d) Shuttle-like cells positively identified 7 days after cell transplantation in cytoplast particles (blue) and nucleus (red). (Prussian blue stain; magnification in c, x100; magnification in d, x200.)

 
The Prussian blue-stained section showed distribution of iron-containing cells mainly in portal triad regions and hepatocyte injury areas, suggesting the presence of transplanted Fe2O3-PLL–labeled BMSCs. These cells showed amorphous shapes with positive iron staining in their cytoplasms and tended to localize along sinusoids (Fig 10). By contrast, Prussian blue positively stained cells were absent in group 2 control rats implanted with unlabeled BMSCs. The group 1 iron-positive cell count in the liver at 3 hours and 3, 7, and 14 days after injection was 67 ± 14.80, 56 ± 13.00, 32 ± 16.37, and 14 ± 5.29, respectively. Quantitative analysis of the BMSC count in the liver from group 1 at different time points demonstrated a significant decrease on days 7 and 14 after injection compared with 3 hours after injection (P < .05).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
 References
 
BMSCs are clonogenic, nonhematopoietic stem cells and are able to differentiate into multiple mesoderm type (eg, osteoblasts, chondrocytes, and endothelial cells) and nonmesoderm type (eg, neuronal-like cells) lineages (29). Studies have demonstrated the intrinsic plasticity of BMSCs for differentiation into hepatocyte-like phenotypes in vitro (5–7). In vivo differentiation of BMSCs into hepatocytes induced by hepatic microenvironment has also been observed (8,9).

There are two topographic approaches for hepatic cell transplantation, orthotopic and ectopic. For ectopic transplantation, such as in the peritoneal cavity, there is a large capacity for indwelling transplanted cells. However, liver cell necrosis occurred soon after transplantation and could not be prevented as a result of attachment of liver cells to microcarriers, suggesting insufficient metabolic support, especially in acute hepatic failure (30).

On the contrary, orthotopic transplantation appears advantageous as a result of the presence of hepatotropic extracellular matrices, the availability of hepatic growth factors, and opportunities for interaction with other liver cells (17). Theoretically, an adult liver can accommodate transplanted cells, up to 70% in addition to the mass of host hepatocytes, in its vascular space.

There are two major approaches for hepatic orthotopic cell transplantation: direct portal vein (transportal) injection and intrasplenic injection. Transportal injection may cause embolization resulting from cell clots in the intrahepatic portal vein and pulmonary artery, resulting in refractory shock. In intrasplenic transplantation, the majority of injected hepatocytes can migrate gradually into the liver by means of the portal venous system, where they integrate with the hepatic parenchyma rather than remaining as intravascular emboli (17).

Repeated transplantation of adequate cell numbers may augment liver repopulation and prevent temporary portal hypertension. In rodents with a superficially positioned spleen, a large fraction of transplanted hepatocytes could be conveniently deposited in hepatic sinusoids through the splenohepatic blood drainage (31–33). All 12 recipient rats with intrasplenic transplantation of BMSCs survived our study without operational mortality before sacrifice, proving favorably the latter approach.

Using the nuclear tracing method, Gupta et al (32) found that the majority of intrasplenically transplanted hepatocytes were able to migrate to the liver parenchyma. However, no results about the distribution of BMSCs after intrasplenic transplantation have been reported, to our knowledge. Different surface markers, homing factors, and size-structure relationships with BMSCs may make their intrahepatic distribution dissimilar to that of hepatocytes.

MR imaging appears most promising for dynamically monitoring in vivo cell migration after transplantation (34). Among MR tracing agents, superparamagnetic iron oxide particles have been most extensively studied in preclinical and clinical research (35–38). Previous in vitro studies have demonstrated that MSCs could be effectively labeled with synthesized iron oxide nanoparticles without substantially changing the viability of labeled cells. The suspension of labeled MSCs could be identified by using a standard 1.5-T MR imager (14). In our study, migration and retention of rat BMSCs after intrasplenic transplantation were demonstrated by using in vivo MR imaging. The applied magnetic labeling technique showed no toxic effects to cells (14,26–28). Three hours after transplantation, the BMSCs migrated to the liver and caused a remarkable decrease in MR SI. Histologic analysis proved that grafted BMSCs initially enter liver sinusoids and are then incorporated into the liver parenchyma, primarily in periportal and injured areas.

In the study with cellular samples, R2* was shown to be dependent on iron oxide concentration alone (39). The sensitivity of R2* to iron oxide–loaded cells was found to be much greater than that of R2 and R1. In addition, its sensitive magnetism effect makes the signal-dropped area larger than the actual cell mass volume, which augments detectability of the magnetic marker (39). In our study, we used an R2*-based imaging method (T2*-weighted MR) to depict accumulated cells as pronounced local SI loss, because it is clinically feasible. However, R2*-based cellular quantification remains complicated in vivo, especially for longitudinal studies (40).

In a previous in vitro MR study, significant SI changes were observed in labeled MSCs at T2- and T2*-weighted sequences (14). The percentage change of SI on T2-weighted images was the greatest. However, T2*-weighted images are vulnerable to susceptibility artifacts, particularly from air-tissue interface, and the negative contrast enhancement may be confounded by other causes of low SI (40). Nevertheless, T1-weighted MR imaging of the corresponding sections may assist with distinguishing and evaluating anatomic structures.

CNRs on T2*-weighted images decreased substantially in the liver 3 hours after injection of BMSCs and gradually returned to the levels found either in control rat livers or before injection. The gradual increase of SI in the liver may be the result of: (a) mobilization of labeled cells out of the liver, (b) dilution of intracellular iron as a result of cell division (41), (c) release of iron out of the cells, or (d) cell death resulting from immune rejection.

A previous study (14) showed that MSCs did not rapidly divide because they needed about 7 days for one in vitro experiment passage. Therefore, it is unlikely that BMSCs underwent several divisions in the liver within 14 days, which would have resulted in a dilution of the iron from the cells. Furthermore, if intracellular iron was released from the cell, it would be subjected to phagocytosis by Kupffer cells and made visible at histologic analysis. However, we did not find Kupffer cells by using Prussian blue staining. Previous studies (42,43) also showed that stem cells home to the liver soon after intravenous administration and gradual mobilization of cells out of the liver. So, the gradual SI increase might be caused by mobilization of cells out of the liver and cellular decease owing to immune rejection.

In most MR imaging studies of transplanted stem cells, cells were implanted locally, for example, in the porcine heart (44,45) and the rat brain and spine (13,46,47), which provided a high concentration of labeled cells. This high concentration is reflected at T2*-weighted imaging as a low SI spot in otherwise unaffected SI of the receiving tissue. Yet, cell migration was limited to a few millimeters per week, and this limitation rendered cell therapy of an entire organ problematic. In the treatment of diffuse diseases, especially diseases of liver, it would be beneficial for grafted stem cells to distribute through the entire organ, which is an important aspect of cell therapy. Since BMSCs dispersed in the liver with an obviously lower concentration than that locally delivered to the heart, brain, and spine, the MR tracing of labeled cells in the liver could be different than that of the heart and the central nervous system (40). Thus, we used in vivo MR imaging to depict and track labeled MSCs in the liver. In our in vivo study, magnetically labeled stem cells underwent intrasplenic transplantation and were successfully detected in the liver by measuring SI decrease of the whole organ.

There were some limitations to our study. First, the number of animals in our study was small. Second, the follow-up was limited to 14 days; therefore, the long-term fate of the cells and the superparamagnetic iron oxide labeling is still unknown. Third, only the liver was imaged as a target organ owing to coil sensitivity constraints. Imaging of other organs, such as the spleen, might enable acquisition of additional useful information. And finally, owing to the difficulty of in vitro culturing of hepatocytes, the labeling, migration, and tracing of BMSCs were not compared with those of hepatocytes.

In conclusion, intrasplenic transplantation is a viable way to seed cells throughout the target liver, and this process holds potential for application in future cell therapy protocols. Magnetic labeling and MR tracing of BMSCs have utility for depiction and evaluation of the transplantation procedures. The combination of the techniques provides a means to deliver cells and immediately verify whether the cells have dispersed throughout the target organ.


   ADVANCES IN KNOWLEDGE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
References
 


   FOOTNOTES
 

Abbreviations: BMSC = bone MSC • CNR = contrast-to-noise ratio • MSC = mesenchymal stem cell • PLL = poly-L-lysine • SI = signal intensity

Author contributions: Guarantors of integrity of entire study, S.J., G.J.T., Y.N.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, S.J., G.J.T., H.L., Y.Z., A.Z., Y.N.; experimental studies, S.J., G.J.T., H.L., Y.Z., A.Z.; statistical analysis, S.J., G.J.T., Y.Z., A.Z.; and manuscript editing, all authors.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 

  1. Allen KJ, Soriano HE. Liver cell transplantation: the road to clinical application. J Lab Clin Med 2001;138:298–312.[CrossRef][Medline]
  2. Petersen BE, Bowen WC, Patrene KD, et al. Bone marrow as a potential source of hepatic oval cells. Science 1999;284:1168–1170.[Abstract/Free Full Text]
  3. Lagasse E, Connors H, Al-Dhalimy M, et al. Purified hematopoietic stem cells can differentiate into hepatocytes in vivo. Nat Med 2000;6:1229–1234.[CrossRef][Medline]
  4. Avital I, Inderbitzin D, Aoki T, et al. Isolation, characterization, and transplantation of bone marrow-derived hepatocyte stem cells. Biochem Biophys Res Commun 2001;288:156–164.[CrossRef][Medline]
  5. Schwartz RE, Reyes M, Koodie L, et al. Multipotent adult progenitor cells from bone marrow differentiate into functional hepatocyte-like cells. J Clin Invest 2002;109:1291–1302.[CrossRef][Medline]
  6. Wang PP, Wang JH, Yan ZP, et al. Expression of hepatocyte-like phenotypes in bone marrow stromal cells after HGF inductions. Biochem Biophys Res Commun 2004;320:712–716.[CrossRef][Medline]
  7. Lee KD, Kuo TK, Whang-Peng J, et al. In vitro hepatic differentiation of human mesenchymal stem cells. Hepatology 2004;40:1275–1284.[CrossRef][Medline]
  8. Yamamoto N, Terai S, Ohata S, et al. A subpopulation of bone cells depleted by a novel antibody, anti-Liv8, is useful for cell therapy to repair damaged liver. Biochem Biophys Res Commun 2004;313:1110–1118.[CrossRef][Medline]
  9. Sato Y, Araki H, Kato J, et al. Human mesenchymal stem cells xenografted directly to rat liver are differentiated into human hepatocytes without fusion. Blood 2005;106:756–763.[Abstract/Free Full Text]
  10. Bulte JW, Douglas T, Witwer B, et al. Monitoring stem cell therapy in vivo using magnetodendrimers as a new class of cellular MR contrast agents. Acad Radiol 2002;9(suppl 2):S332–S335.[CrossRef][Medline]
  11. Soriano HE, Lewis D, Legner M, et al. The use of DiI-marked hepatocytes to demonstrate orthotopic, intrahepatic engraftment following hepatocellular transplantation. Transplantation 1992;54:717–723.[Medline]
  12. Bulte JW, Zhang SC, van Gelderen P, et al. Magnetically labeled glial cells as cellular MR contrast agents. Acad Radiol 2002;9(suppl 1):S148–S150.[CrossRef][Medline]
  13. Modo M, Cash D, Mellodew K, et al. Tracking transplanted stem cell migration using bifunctional, contrast agent–enhanced magnetic resonance imaging. NeuroImage 2002;17:803–811.[CrossRef][Medline]
  14. Ju S, Teng G, Mao X, et al. In vitro labeling and MR imaging of mesenchymal stem cells from human umbilical cord blood. Magn Reson Imaging 2006;24(6):611–617.[CrossRef][Medline]
  15. Ponder KP, Gupta S, Leland F, et al. Mouse hepatocytes migrate to liver parenchyma and function indefinitely after intrasplenic transplantation. Proc Natl Acad Sci U S A 1991;88:1217–1221.[Abstract/Free Full Text]
  16. Gupta S, Rajvanshi P, Lee CD. Integration of transplanted hepatocytes into host liver plates demonstrated with dipeptidyl peptidase IV-deficient rats. Proc Natl Acad Sci U S A 1995;92:5860–5864.[Abstract/Free Full Text]
  17. Rajvanshi P, Kerr A, Bhargava KK, Burk RD, Gupta S. Studies of liver repopulation using the dipeptidyl peptidase IV-deficient rat and other rodent recipients: cell size and structure relationships regulate capacity for increased transplanted hepatocyte mass in the liver lobule. Hepatology 1996;23:482–496.[Medline]
  18. Ma M, Zhang Y, Yu W, Shen H, Zhang H, Gu N. Preparation and characterization of magnetite nanoparticles coated by amino silane. Coll Surf 2003;212:219–226.[CrossRef]
  19. Ma M, Wu Y, Zhou J, Sun Y, Zhang Y, Gu N. Size-dependence of specific power absorption of Fe3O4 particles in AC magnetic field. J Magn Mater 2003;268:33–39.
  20. Prockop DJ. Marrow stromal cells as stem cells for nonhematopoietic tissues. Science 1997;276:71–74.[Abstract/Free Full Text]
  21. Aubin JE. Osteoprogenitor cell frequency in rat bone marrow stromal populations: role for heterotypic cell-cell interactions in osteoblast differentiation. J Cell Biochem 1999;72:396–410.[CrossRef][Medline]
  22. Sanchez-Ramos J, Song S, Cardozo-Pelaez F, et al. Adult bone marrow stromal cells differentiate into neural cells in vitro. Exp Neurol 2000;164:247–256.[CrossRef][Medline]
  23. Dezawa M, Kanno H, Hoshino M, et al. Specific induction of neuronal cells from bone marrow stromal cells and application for autologous transplantation. J Clin Invest 2004;113:1701–1710.[CrossRef][Medline]
  24. Recknagel RO, Glende EA, Dolak JA, et al. Mechanism of CCl4 toxicity. Pharmacol Ther 1989;43:139–154.[CrossRef][Medline]
  25. Wolff SD, Balaban RS. Assessing contrast on MR images. Radiology 1997;202:25–29. [Published correction appears in Radiology 1997;202:880–881.][Abstract/Free Full Text]
  26. Liu L, Tang M, Gu N, et al. Study on acute and chronic toxicity to rats of nanoparticles of Fe2O3 coated with glutamic acid [in Chinese]. J Environ Occup Med 2004;21:430–433.
  27. Li Q, Tang M, Ban T, Ma M, Gu N. Study on cytotoxicity and oxidative effects of different sizes of Fe2O3 nanoparticles on CHL cells in vitro [in Chinese]. Chin J Public Health 2005;21:589–591.
  28. Yin Q, Liu L, Gu N, et al. Determining the biodistribution of nano-Fe2O3-Glu in mice by 59Fe tracer and preparation [in Chinese]. J Med Postgrad 2005;18:312–317.
  29. Kassem M, Kristiansen M, Abdallah BM, et al. Mesenchymal stem cells: cell biology and potential use in therapy. Basic Clin Pharmacol Toxicol 2004;95:209–214.[CrossRef][Medline]
  30. Henne-Bruns D, Kruger U, Sumpelmann D, Lierse W, Kremer B. Intraperitoneal hepatocyte transplantation: morphological results. Virchows Arch A Pathol Anat Histopathol 1991;419:45–50.[CrossRef][Medline]
  31. Gupta S, Yerneni PR, Vemuru RP, Lee CD, Yellin EL, Bhargava KK. Studies on the safety of intrasplenic hepatocyte transplantation: relevance to ex vivo gene therapy and liver repopulation in acute hepatic failure. Hum Gene Ther 1993;4:249–257.[Medline]
  32. Gupta S, Lee CD, Vemuru RP, Bhargava KK. 111Indium labeling of hepatocytes for analysis of short-term biodistribution of transplanted cells. Hepatology 1994;19:750–757.[CrossRef][Medline]
  33. Gupta S, Aragona E, Vemuru RP, Bhargava KK, Burk RD, Chowdhury JR. Permanent engraftment and function of hepatocytes delivered to the liver: implications for gene therapy and liver repopulation. Hepatology 1991;14:144–149.[CrossRef][Medline]
  34. Unger EC. How can superparamagnetic iron oxides be used to monitor disease and treatment? Radiology 2003;229:615–616.[Free Full Text]
  35. Yeh TC, Zhang W, Ildstad ST, Ho C. In vivo dynamic MRI tracking of rat T-cells labeled with superparamagnetic iron-oxide particles. Magn Reson Med 1995;33:200–208.[Medline]
  36. Josephson L, Tung CH, Moore A, Weissleder R. High-efficiency intracellular magnetic labeling with novel superparamagnetic-Tat peptide conjugates. Bioconjug Chem 1999;10:186–191.[CrossRef][Medline]
  37. Daldrup-Link HE, Rudelius M, Oostendorp RA, et al. Targeting of hematopoietic progenitor cells with MR contrast agents. Radiology 2003;228:760–767.[Abstract/Free Full Text]
  38. Frank JA, Miller BR, Arbab AS, et al. Clinically applicable labeling of mammalian and stem cells by combining superparamagnetic iron oxides and transfection agents. Radiology 2003;228:480–487.[Abstract/Free Full Text]
  39. Bowen CV, Zhang X, Saab G, Gareau PJ, Rutt BK. Application of the static dephasing regime theory to superparamagnetic iron-oxide loaded cells. Magn Reson Med 2002;48:52–61.[CrossRef][Medline]
  40. Bos C, Delmas Y, Desmouliere A, et al. In vivo MR imaging of intravascularly injected magnetically labeled mesenchymal stem cells in rat kidney and liver. Radiology 2004;233:781–789.[Abstract/Free Full Text]
  41. Arbab AS, Jordan EK, Wilson LB, Yocum GT, Lewis BK, Frank JA. In vivo trafficking and targeted delivery of magnetically labeled stem cells. Hum Gene Ther 2004;15:351–360.[CrossRef][Medline]
  42. Gao J, Dennis JE, Muzic RF, Lundberg M, Caplan AI. The dynamic in vivo distribution of bone marrow-derived mesenchymal stem cells after infusion. Cells Tissues Organs 2001;169:12–20.[CrossRef][Medline]
  43. Aicher A, Brenner W, Zuhayra M, et al. Assessment of the tissue distribution of transplanted human endothelial progenitor cells by radioactive labeling. Circulation 2003;107:2134–2139.[Abstract/Free Full Text]
  44. Kraitchman DL, Heldman AW, Atalar E, et al. In vivo magnetic resonance imaging of mesenchymal stem cells in myocardial infarction. Circulation 2003;107:2290–2293.[Abstract/Free Full Text]
  45. Hill JM, Dick JA, Raman VK, et al. Serial cardiac magnetic resonance imaging of injected mesenchymal stem cells. Circulation 2003;108:1009–1014.[Abstract/Free Full Text]
  46. Bulte JW, Zhang S, van Gelderen P, et al. Neurotransplantation of magnetically labeled oligodendrocyte progenitors: magnetic resonance tracking of cell migration and myelination. Proc Natl Acad Sci U S A 1999;96:15256–15261.[Abstract/Free Full Text]
  47. Hoehn M, Kustermann E, Blunk J, et al. Monitoring of implanted stem cell migration in vivo: a highly resolved in vivo magnetic resonance imaging investigation of experimental stroke in rat. Proc Natl Acad Sci U S A 2002;99:16267–16272.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2443061290v1
245/1/206    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ju, S.
Right arrow Articles by Ni, Y.
Right arrow