Cell Technologies in Biology and Medicine, No. 3, November, 2012
145
Methods of Detection of Mesenchymal Stem Cells in the Kidneys during Therapy of Experimental Renal Pathologies
E. Yu. Plotnikov1, N. V. Pul’kova1, D. N. Silachev1, V. N. Manskikh2, T. G. Khryapenkova1, D. B. Zorov1, and G. T. Sukhikh3 Translated from Kletochnye Tekhnologii v Biologii i Meditsine, No. 3, pp. 152-159, July, 2012 Original article submitted June 5, 2012
We studied the possibility of using different methods for identification of mesenchymal multipotent stromal cells in the renal parenchyma under conditions of experimental thermal ischemia of the kidneys and acute pyelonephritis. In vivo and in vitro methods of identification of mesenchymal multipotent stromal cells by magnetic resonance imaging and immunological and immunohistochemical methods were compared. Labeling of stem cells with iron-containing particles followed by their histological identification and immunohistochemical staining with species-specific antibodies were the most informative methods. Active migration of the cells to the renal tissue was detected by these methods in experimental acute pyelonephritis with inflammation foci. Key Words: stem cells; ischemia of the kidney; pyelonephritis; migration; magnetic resonance imaging Various aspects of using cell technologies for the therapy of renal pathologies of different etiology and pathogenesis are now extensively studied. The efficiency of transplantation of mesenchymal multipotent stromal cells (MMSC) in the therapy of acute and chronic renal failure [1,2] and pyelonephritis has been proven in experimental and clinical studies. However, the mechanism of nephroprotective effect of SC is still poorly understood. Some concepts of nephroprotective activity of MMSC are discussed, of them the replacement, immunomodulating, and paracrine mechanisms are thought to be most relevant, but realization of these mechanisms requires SC delivery to the damaged organ or tissue. SC can be guided to the target tissue by the gradient of signal molecules due to the so-called homing effect [4]. For the analysis of this migration, 1 A. N. Belozerskii Institute of Physicochemical Biology, M. V. Lomonosov Moscow Medical University; 2Faculty and Bioengineering and Bioinformatics, M. V. Lomonosov Moscow State University; 3V. I. Kulakov Research Center of Obstetrics, Gynecology, and Perinatology, Ministry of Health and Social Development of the Russian Federation, Moscow, Russia. Address for correspondence: zorov@ genebee.msu.ru. D. B. Zorov
MMSC introduced into the body should be traced and identified in the target organ throughout a long (ideally, life-long) period. The methods of studies of cell distribution in the body can be divided into intravital and postmortem. Each of them requires techniques of identification and discrimination of transplanted cells into the recipient tissues. To this end, exogenous markers associated with transplanted cells or antigenic determinant of these cells are used. The ideal methods for SC tracing should be sensitive enough for visualization of a single cell and allow precise evaluation of the number and localization of cells in any anatomical structure. The exogenous markers used for these purposes should be biologically compatible, safe, and nontoxic for the cells. Here we studied different methods of targeted migration of transplanted MMSC into the kidney and their distribution in the tissue in various renal pathologies. The model of thermal ischemia of the kidney accompanied by extensive death of nephron cells and the model of pyelonephritis accompanied by acute inflammatory response were used. The choice of these
0007-4888/12/15410145 © 2012 Springer Science+Business Media New York
146
Cell Technologies in Biology and Medicine, No. 3, November, 2012
models was dictated by different damaging mechanisms. It can be hypothesized that MMSC migration into the kidney will differ under these conditions and will allow us to compare different methods of MMSC identification. The aim of this study was a comparative analysis of different methods of MMSC tracing and identification during their migration to the kidney under conditions of experimental pathology.
justed to a required concentration and injected to experimental animals. Evaluation of the state of cell mitochondria Mitochondrial transmembrane potential was evaluated using a potential-sensitive probe TMRE (red fluorescence). The differences between the extra- and intramitochondrial concentrations of TMRE allow clear-cut visualization of normally energized mitochondria. The cell culture was stained with TMRE in DMEM/F12 with 20 mM HEPES. The cells were incubated for 30 min at 37oC in a medium containing 200 nM TMRE. Then the culture was washed with the medium from dye excess. Labeling with iron oxide microparticles. For labeling with iron oxide microparticles (IOMP), the cells were grown in flasks (105/ml) in DMEM and commercial IOMP preparation (SiMAG, 0.75 μ; chemicell) was added to a concentration of 1 μl/ml. The cells were incubated with SiMAG for 24 h at 37oC and 5% CO2. After incubation, the cells were washed with 3 portions of 0.9% NaCl (PanEko), dissociated with 0.25% trypsin-EDTA, centrifuged at 100g, adjusted to the required concentration, and injected to animals. For verification of cells loading with IOMP, an aliquot of cells was fixed on a glass and stained by the method of Perls to confirm the presence of iron in the cytoplasm. Flow cytofluorometry. The contents of calceinlabeled MMSC in kidney homogenate in 1 and 24 h after injection of cells were compared on a Beckman Coulter flow cytometer. Unlabeled MMSC and MMSC loaded with calcein 1 h before cytofluorometry were used as negative and positive control, respectively. Fluorescence was excited with a 488 nm laser and measured at 510-530 nm. Magnetic resonance imaging. Magnetic resonance imaging (MRI) was carried out on a BioSpec 70/30 tomograph at 7 T magnetic field power (Center of Magnetic Tomography and Spectroscopy, M. V. Lomonosov Moscow State University). The study was performed under general anesthesia (300 mg/kg chloral hydrate, intraperitoneally) on days 3 and 14 after cell injection. Multispin echo T2-weighted imaging (T2-WI) was used for identification of IOMP-labeled MSC. A linear transmitting coil with an inner diameter of 72 mm and a surface receiving coil for detection of the radiofrequency signal from the rat brain were used. For obtaining T2-WI, we used RAPE (Rapid Acquisition with Relaxation Enhancement)-based pulse sequence with the following parameters: TR=6000 msec, TE=63.9 msec, 0.5 mm slice thickness, matrix size 2568384, resolution 0.164×0.164 mm/μ. The total time of scanning was 4 min 48 sec. Immunohistochemical staining. Kidney sections were washed with PBS, treated with 0.2% Triton
MATERIALS AND METHODS MMSC culturing. MSC were isolated and characterized at V. I. Kulakov Research Center of Obstetrics, Gynecology, and Perinatology. The cells were washed out from BM with a syringe with DMEM containing 2 mM EDTA as the anticoagulant. Then, the cell suspension was layered onto Ficoll-urografin density gradient (1.077 g/ml) and centrifuged at 200g for 30 min. The interphase mononuclear cell fraction was collected, resuspended in the medium, and re-centrifuged at 150g for 5 min. The pellet was resuspended in complete nutrient medium DMEM/F12 (1:1) supplemented with 10% embryonic calf serum and 0.02% gentamicin and transferred to flasks (25 cm2, Corning). After MMSC adhesion to the plastic, nonadherent cells were removed and the adherent culture was grown until confluence. The number of viable cells was counted by staining with trypan blue and propidium iodide; cells with viability >90% were used for culturing. Modeling of acute pyelonephritis. Experiments were carried out on outbred albino female rats weighing 180-200 g maintained in a vivarium at 12-h illumination regimen with free access to food and water. The animals were anesthetized with isofluorane and a suspension of cultured fecal bacteria (5 ml/kg, 108 CFU/ml, the content of E. coli 90%) was administered through a urethral catheter. Controls received 0.9% NaCl instead of bacterial suspension. In 3 days after administration of bacterial suspension, MMSC were injected into the jugular vein in a dose of 1.5×107 cells/kg. On day 7 of pyelonephritis, the rats were sacrificed and the kidneys were collected for histological analysis. Modeling of renal ischemia. Ischemia of the left kidney (90 min) was modeled as described previously [6] and simultaneously with right-sided nephrectomy. MMSC labeling with fluorescent dye calcein. For labeling the cytoplasm content, the cells were labeled with Calcein-AM. To this end, the cells were incubated for 30 min in a medium with 2.5 μM Calcein-AM (Molecular probes) at 37oC, washed twice in culture medium, and dissociated with 0.25% trypsin-EDTA (PanEko). The cells were centrifuged at 100g and ad-
E. Yu. Plotnikov, N. V. Pul’kova, et al.
X-100 in PBS for 60 min at 25oC, than twice washed with PBS, and blocked with 0.5% BSA (PBS-BSA) for 60 min at 25oC. The sections were incubated for 12 h at 4oC with mouse monoclonal antibodies to human nuclei (Chemicon) diluted 1:50, washed with PBSBSA (3×15 min), and incubated for 1 h at 25oC with anti-mouse antibodies conjugated with FITC. Then the cells were washed with PBS-BSA (3×15 min), embedded into 50% glycerin in PBS, and fixed with a lacquer. Fluorescence was detected with a confocal microscope. Fluorescence was excited with a 488-nm argon laser and recorded at 505-530 nm.
RESULTS MMSC labeling with fluorescent dye calcein. Calcein-AM is used for staining of viable cells. Cleavage with cytoplasmic esterase releases the fluorescent form of calcein. In this state, the dye cannot leave the cell. For studying migration and integration of transplanted cells into renal structures after ischemia of the kidney,
147 we injected calcein-loaded MMSC into the kidney. Examination of kidney sections after 24 h revealed labeled MMSC at the site of injection and adjacent areas (Fig. 1, a). Labeled cells were detected only in the interstitial space of the kidney tissue even at a great distance from the site of injection, but not in the tubular epithelium (Fig. 1, a). Double staining of vital sections of the kidney with mitochondrial dye TMRE and Calcein AM showed that considerable number of injected MMSC retained mitochondrial potential (Fig. 1, b). This suggests that the greater part of transplanted cells is intact and functionally active. However, many calcein-stained cells were not stained with TMRE, i.e. were dead or their mitochondrial functions were impaired. The absence of calcein fluorescence in cells lining the renal tubules probably suggests that MMSC do not incorporate into epithelial lining of damaged tubules, at least at early terms after injection. At later terms after transplantation (7 days), calcein-labeled MMSC were still seen in the kidney, but their distribution looked more diffuse. No cell groups
Fig. 1. Detection of MMSC by confocal microscopy. IOMP-labeled MMSC were injected 90 min after thermal ischemia of the kidney. Additional staining of the sections with fluorescent probe TMRE (red fluorescence). a) vital slices of the kidney on day 1 after MMSC transplantation: distribution of injected cells (green fluorescence) along the kidney interstitium. Arrow shows the site of injection; b) staining of viable MMSC with both dyes yields yellow coloration after superposition (arrows); c) vital slices of the kidney in 7 days after MMSC injection; d) calcein-stained cell in the renal tubule (arrow).
148
Cell Technologies in Biology and Medicine, No. 3, November, 2012
were seen at the site of injection. The number of cells per sections was considerably lower (Fig. 1, c) than on day 1. MMSC migrated far from the site of injection practically all over the kidney parenchyma, which was confirmed by serial sections of the kidneys. Mitochondrial potential of the majority of calcein-labeled cells was relatively low, but comparable with that in other tubular cells. This suggests that donor cells retained functional mitochondria, i.e. were viable at the specified terms of the experiment. Most of calcein-labeled cells were detected in the interstitium, but some cells were also found in tubules (Fig. 1, d).
For quantitative detection of migrated MMSC in the whole kidney, the method of flow cytofluorometry was used. As was previously described, the calceinlabeled cells were injected into the parenchyma of ischemic kidney (5×106 cells/kg, 106 cells per kidney). In 1 and 24 h, the kidney tissue was completely dissociated with type II collagenase and the suspension was analyzed on a flow cytofluorometer. In 1 h, calceinlabeled cells constituted about 1% of all cells (Fig. 2, c), but in 24 h calcein-labeled cells were not detected. Thus, about 105 MMSC were detected 1 h after their injection into the kidney (i.e. 10-fold lower than the
Fig. 2. Evaluation of the content of calcein-labeled cells in kidney homogenate after their injection into the parenchyma after modeling kidney ischemia by the method of flow cytometry. Negative control (unlabeled MMSC; a) and labeled MMSC before (b) and 1 (c) and 24 h (d) after injection into the kidney.
E. Yu. Plotnikov, N. V. Pul’kova, et al.
149
Fig. 3. Identification of IOMP-labeled MMSC in the kidney by MRI and histochemical staining. a) IOMP-labeled MMSC stained by the method of Perls in culture; b) magnetic resonance images of phantoms consisting of a set of tubes with IOMP-labeled MMSC suspension in 10 ml gelatin of different concentrations: 1) cell-free gelatin solution; 2) gelatin solution with unlabeled MMSC; 3-7) suspension of IOMP-labeled MMSC prepared by the method of serial 2-fold dilutions and containing 106, 5105, 2.5105, 1.25104, and 6.25102 cells, respectively; c-e) magnetic resonance images of rat kidneys after 90-min thermal ischemia of the kidney and injection of IOMP-labeled MMSC into the kidney parenchyma: iron-labeled cells are seen as a dark area (arrow); c) day 3 after transplantation; d) 2 weeks after transplantation; e) intact kidneys; f-g) identification of IOMP-labeled MMSC on histological slices of the kidney by Perls staining in 7 days after pyelonephritis induction (IOMP-labeled MMSC are shown by arrows); f) intact kidney (400); g) kidney with pyelonephritis (200).
injected dose), which characterizes this method as low sensitive and low informative. The limitations of this method can be related to distraction of some transplanted cells during enzymatic dissociation or incomplete dissociation of the kidney tissue, due to which the cells located within large cell agglomerates are not recorded during cytometry. Similar approach was used for evaluation of migration of calcein-labeled MMSC into the kidney after their intravenous injection to animals with modeled acute pyelonephritis. The cells were transported 24 h after pyelonephritis modeling. In this case, crude
homogenate of the kidney prepared in a glass-Teflon Potter homogenizer was centrifuged at 10,000g for removal of all cell components and integral intensity of calcein fluorescence in the extract was measured. Calcein content in the tissue reflecting the number of migrated MMSC was analyzed in 24 h after cell transplantation. The intensity of calcein fluorescence in the kidney homogenates of control rats receiving labeled MMSC was 72.9±31.4 rel. units and corresponded to the level of fluorescence receiving unlabeled MMSC. After pyelonephritis modeling, the intensity of calcein fluorescence increased by 30 times to 2366.8±718.5
150
Cell Technologies in Biology and Medicine, No. 3, November, 2012
Fig. 4. Detection of MMSC by immunohistochemical staining with antibodies to human nuclei (green fluorescence). Confocal microscopy of fixed kidney sections 7 days after 90-min thermal ischemia and injection of MMSC into the parenchyma. a) site of cell injection: MMSC are detected in the interstitium of the kidney; b) localization of injected MMSC around the renal tubule.
rel. units, which attested to effective migration of MMSC into the kidney during inflammation. Calcein-AM is a non-toxic compound and can be used for cell tracing in short-term experiments. The intensity of its reuptake by recipient cells (after death of transplanted cells and due release) is low, which was confirmed by low accumulation of the dye in the kidneys of intact animals. The main advantage of using vital slices of the kidney for identification of transplanted calcein-labeled MMSC is the possibility of identification of viable cells by treatment with TMRE accumulating in only functionally active mitochondria. However, the technique of vital slice preparation is laborious and requires confocal microscopy because of heterogeneity and great thickness of the slices. Identification of MMSC in the kidney homogenates by flow cytofluorometry and measurement of total fluorescence intensity allow evaluation of the net content of cells in the organ. Flow cytofluorometry allows counting of each single cell. However, this approach is characterized by low sensitivity. Total fluorescence intensity only indirectly reflects the number of migrating MMSC; otherwise, construction of calibration curves with the use of large animal groups is required. MMSC labeling with iron oxide microparticles. Intravital evaluation of MMSC migration into the kidney at different terms after transplantation was evaluated by T2-weighed MRI. For MMSC identification by MRI, the cells were preincubated with IOMP over 24 h. Iron microparticles exhibit paramagnetic pro-
perties and are used as contrast agents for MRI. The absorption of IOMP by MMSC culture was verified by staining for Fe3+ ions by the method of Perls (Fig. 3, a). For evaluation of method sensitivity, phantoms were made of 10% gelatin with addition of IOMPlabeled MMSC by the method of serial 2-fold dilutions (Fig. 3, b). The maximum dilution allowing cell detection was 2.5×105 cells per 10 ml (Fig. 3, b). IOMP-labeled MMSC were injected into the kidney parenchyma 1 h after ischemia. MRI was performed on days 3 and 14. The cells on magnetic resonance images were seen as hypointensive (dark) areas (Fig. 3, c-e). The size and location of this area remained unchanged over 2 weeks. This suggests that the bulk of cells did not migrate or migrated at a short distance; high elimination capacity of the kidney tissues excludes the possibility of iron accumulation in the renal epithelial cells. It is known that the dye used for SC staining is rapidly absorbed by cells and then degraded with Fe3+ release that is rapidly metabolized. Therefore, the zones detected by MRI do represent groups of injected cells, but not kidney cells that absorbed the preparation. However, it should be noted that MRI even with technical characteristics used in the present study does not provide resolution comparable with the cell size. That is why cell migration into the kidney was studied on the model pyelonephritis with intravenous transplantation IOMP-labeled cells followed by histochemical staining of kidney slices for Fe3+ by the Perls method based on Berlin blue
E. Yu. Plotnikov, N. V. Pul’kova, et al.
formation. Analysis of histological sections showed numerous cell clusters with typical blue staining in all layers of the kidney of rats with pyelonephritis receiving IOMP-labeled MMSC (19.5±3.7 cells per field of view). The maximum accumulation of these cells coincided with sites of inflammatory infiltration. In healthy animals, only solitary cells were identified on kidney sections (2.3±0.4; Fig. 3, f). MRI performed on the next day after transplantation revealed no IOMPlabeled cells in the kidney, which attested to low sensitivity of this method. Thus, IOMP labeling of MMSC is a universal method allowing both intravital tracing and postmortem identification of cells in the kidney tissue using simple histochemical staining. However, MRI is a low-sensitive method requiring high content of labeled cells in the target organ. Previous studies showed that accumulation of iron particles in concentrations sufficient for cell identification produced no toxic effect on the cell and had no effect on their proliferation [3,5]. Detection of xenogeneic MMSC by immunological methods. Another method confirming the presence of transplanted cells in the kidney and more precisely determining their localization is staining of fixed kidney slices with antibodies to human nuclei. This method is based on injection of autologous SC (in our experiments, human SC were injected to rats) with their subsequent immunochemical staining for antigens expressed only by transplanted cells. The kidney subjected by thermal ischemia was isolated in 7 days after cell transplantation and stained with specific antibodies to human nuclear proteins. As is seen on confocal images, the cells were primarily located in the interstitium (Fig. 4, a, b). Thus, the results obtained by using this approach coincided with those obtained in experiments with calcein staining, but this method is simpler and does not require preliminary MMSC loading with a dye and methodically difficult preparation of intravital slices. Moreover, standard histological and immunochemical methods make it possible to prepare, store, and analyze multiple serial sections of the kidney. In the present study, the possibility of using various methods of MMSC identification in the kidney parenchyma after their administration via different routes was demonstrated on the models of hear ischemia of the kidney and acute pyelonephritis.
151 The analyzed methods of tracing MMSC migration into the kidney have some advantages and limitations. For instance, calcein labeling and confocal microscopy of vital sections followed by additional staining with mitochondrial dye TMRE allowed identification of transplanted cells and evaluation of their functional activity. Identification of IOMP-labeled MMSC by MRI was characterized by low sensitivity. The cells were detected only after their injection into the kidney parenchyma on the model of ischemia, whereas no signal from the cells was obtained after their systemic administration on the model of acute pyelonephritis. Other authors reported the presence of the signal only after cell injection into the renal artery [3,5]. These are also potential limitations of tracing of iron-labeled cells by MRI related to possible uptake of iron particles by macrophages, which can lead to misinterpretation of the experimental results. The simplest method of cell identification in the kidney was immunohistochemical staining for species-specific antigens. However, SC from different animal species are required for this approach, which is not always acceptable and available variant. Thus, studies of MMSC migration into the kidney on the models of kidney ischemia and acute pyelonephritis with the use of different methods of MMSC identification revealed differences in targeted migration of transplanted cells into the kidney. The cells most actively migrate in experimental acute pyelonephritis, which can be explained by active production of inflammation-related signal molecules by the kidneys [7].
REFERENCES 1. V. I. Kirpatovskii, A. V. Kazachenko, E. Yu. Plotnikov, et al., Kletoch. Tekhnol. Biol. Med., No. 2, 70-76 (2006). 2. V. I. Kirpatovskii, A. V. Kazachenko, E. Yu. Plotnikov, et al., Kletoch. Tekhnol. Biol. Med., No. 1, 52-57 (2007). 3. C. Bos, Y. Delmas, A. Desmouliere, et al., Radiology, 233, 781-789 (2004). 4. X. Fu, B. Han, S. Cai, et al., Wound Repair. Regen., 17, 185191 (2009). 5. H. Ittrich, C. Lange, F. Togel, et al., J. Magn. Reson. Imaging., 25, 1179-1191 (2007). 6. E. Y. Plotnikov, T. G. Khryapenkova, M. Y. Vyssokikh, et al., Kidney Int., 72, 1493-1502 (2007). 7. A. Ponte, E. Marais, N. Gallay, et al., Stem Cells, 25, 17371745 (2007).