Role of adipose tissue-derived stem cells in the progression of renal disease

Objective: To analyze the role of adipose tissue-derived stem cells in reducing the progression of renal fibrosis. Methods: adipose tissue-derived stem cells were isolated from C57Bl/6 mice and characterized by cytometry and differentiation. Renal fibrosis was established after unilateral clamping of the renal pedicle for 1 hour. Four hours after reperfusion, 2.105 adipose tissue-derived stem cells were administered intraperitoneally and the animals were followed for 24 hours during 6 weeks. In another experimental group, 2.105 adipose tissue-derived stem cells were administered only after 6 weeks of reperfusion, and they were euthanized and studied 4 weeks later. Twenty-four hours after reperfusion, the animals treated with adipose tissue-derived stem cells displayed reduced renal and tubular dysfunction and an increase of the regenerative process. Renal expression of IL-6 and TNF mRNA were decreased in the animals treated with adipose tissue-derived stem cells, while the levels of IL-4, IL-10, and HO-1 were increased, despite the fact that adipose tissue-derived stem cells were not observed in the kidneys via SRY analysis. Results: In 6 weeks, the kidneys of non-treated animals decreased in size, and the kidneys of the animals treated with adipose tissue-derived stem cells remained at normal size and display less deposition of type 1 collagen and FSP-1. The renal protection observed in animals treated with adipose tissue-derived stem cells was followed by a drop in serum levels of TNF-α, KC, RANTES, and IL1a. Treatment with adipose tissue-derived stem cells after 6 weeks, when the animals already displayed established fibrosis, demonstrated an improvement in functional parameters and less fibrosis analyzed by Picrosirius stain, as well as a reduction of the expression of type 1 collagen and vimentin mRNA. Conclusion: Treatment with adipose tissue-derived stem cells may deter the progression of renal fibrosis by modulation of the early inflammatory response, likely via reduction of the epithelial-mesenchymal transition.


INTRODUCTION
Despite all efforts in treating acute renal failure (ARF), this syndrome is still associated with high rates of mortality and morbidity.Additionally, it displays a high incidence and may affect about 7% of all hospitalized patients, with a mortality rate close to 50% in patients in Intensive Care Units (ICUs) (1) .Among the long-term (1 to 10 years) ARF survivors, approximately 12.5% are dialysis-dependent and 19 to 31% develop chronic renal disease (CRD).This scenario may be due to incomplete reparation and persistent tubulointerstitial inflammation, leading to an increase in proliferation of fibroblasts and activation of myofibroblasts, resulting in excessive deposition of extracellular matrix (2) .The epithelial-mesenchymal transition (EMT) has been suggested as the key factor for the development of fibrosis in the progression from ARF to CRD (3)(4)(5).
EMT may be seen as an adaptive response of epithelial cells after chronic stress or lesion.Hypoxia, oxidative stress, and inflammation may produce and release various chemokines and cytokines that attract and direct the influx of inflammatory cells to the tubulointerstitial space.This cellular infiltrate is activated and produces a mix of soluble factors, including proinflammatory agents, profibrotic cytokines, and metallopeptidase-9 matrix (MMP-9) (6) , creating a hostile microenvironment for epithelial and endothelial cells.During the early phase of the inflammatory process in renal fibrosis, cytokines produced by the cell infiltrate such as IL-1, TNF-α, and IFNγ, potentiate tubular EMT triggered by TGF-β, inducing the expression of the TGF-β receptor (7) .TGF-β is the primary profibrotic factor, since it promotes the proliferation of fibroblasts, synthesis of the extracellular matrix, and inhibition of collagenase in multiple organs, events that are characteristic of EMT (8,9) .Activation of NF-kB-dependent TNF-α also stabilizes the Snail transcription factor (a potent EMT inducer) blocking its ubiquitination and providing another molecular link between inflammation and EMT (10,11) .The importance of renal inflammation in initiating and promoting EMT is also manifested by many observations that renal fibrosis is always preceded by and intimately associated with chronic interstitial inflammation (12)(13)(14) .
Adipose tissue-derived stem cells (ADSCs) are adult mesenchymal stem cells (15) that are easily isolated and, when influenced by the extracellular environment, have the capacity, in vitro, to differentiate into other cell types, such adipocytes, myocytes, osteoblasts and neurons.They are known to secrete various growth factors, and therefore, to possess a cytoprotective effect in different lesion models (16) .Gimble et al. demonstrated that when present in damaged tissues, ADSCs can secrete cytokines and growth factors that stimulate the recovery and cleansing of toxic substances released at the site of injury, promoting recovery of the surviving cells (17) .Various studies demonstrated that ADSCs also have the capacity to regulate responses of the immune system (18,19) .They can suppress activation of T-cells by inhibiting expression of cyclin D2 (20) , by inducing regulator T-cells (21) , or by interfering with dentritic cells (21,22) .Additionally, ADSCs may act by means of an anti-apoptotic effect on targetcells, increasing recovery of renal function.
Recently, therapies with stem cells have proved to be a new strategy to reduce the progression of CRD.Semedo et al. demonstrated that the administration of mesenchymal stem cells (MSCs) improved histological and functional parameters, besides reducing fibrosis in a model of serious renal ablation.The renal protection resulting from the administration of MSCs also attenuated the chronic hypoxia, oxidative stress, and inflammatory cytokines that lead to EMT and are present during CRD (23,24) .Several other studies described the beneficial effects of ADSCs administration.Treatment with ADSCs leads to better vascularization (25) , cranial bone regeneration, cardiac wall regeneration, infarcted myocardium function repair, and functional improvement after a cerebrovascular accident.In this study, we proposed that ADSCs may interrupt the progression of renal fibrosis in an animal model of CRD by modulating inflammatory events.

Isolation of ADSCs
Adipose tissue from the inguinal region of the mouse was isolated from male C5Bl/6 mice and digested with 0.075% collagenase IA (Sigma Aldrich, St Louis, MO, USA) at 37ºC for 30 minutes.The cell suspension was filtered through a 70 μm filter (BD Biosciences, San Jose, CA, USA) in order to remove tissue debris.The collagenase was then inactivated with fetal bovine serum (FBS) and the cell suspension was washed with a phosphate buffer (PBS) and centrifuged twice, for 5 minutes, at 260 xg each time.The pellet formed was suspended in 0.84% NH4Cl to remove red blood cells.The cells were then washed and centrifuged twice with PBS and cultivated in plastic jars (TPP, Trasadingen, Switzerland) in low-glucose Dulbecco's modified Eagle medium (DMEM, Gibco, Invitrogen Corporation, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (Emcare, Campinas, SP, Brazil).The cells were cultured for 4 weeks.

ADSCs immunophenotyping
The suspension of ADSCs (1 × 105 cells) was incubated for 40 minutes in saturated concentrations of antibodies CD34-FITC, CD105-FITC, CD73-CD45-PE, and PerCP, and controls (all antibodies were acquired from BD PharMingen Biosciences, San Diego, CA, USA).After three washings, the cells were centrifuged at 200 xg during 5 minutes and the resulting pellet was resuspended in a cold phosphate buffer (PBS).Cellular fluorescence was assessed with the use of a flow cytometer (FACSCanto, Becton Dickinson, Franklin Lakes, NJ, USA).

Animal model of CKD
Female C57BL/6J mice with 8 weeks of age coming from Centro de Desenvolvimento de Modelos Experimentais (CEDEME) of the Universidade Federal de São Paulo (UNIFESP) were submitted to unilateral renal ischemia, as described by Burne-Taney et al (26) .The animals were submitted to abdominal incisions, and their right renal pedicles were dissected.A microvascular clamp (Rocca, São Paulo, SP, Brazil) was placed on the left renal pedicle for 60 minutes, and the animals were maintained at an approximate temperature of 37ºC.The animals were monitored and maintained for 24 hours, and for 6 and 10 weeks before being euthanized.Sham animals were submitted to the same surgical procedure, but without renal artery clamping.All experimental procedures were approved by the Research Ethics Committee of UNIFESP (CEP process # 1915/2008).

Treatment with ADSCs
Four hours after the surgical procedure, one group of animals received ADSCs intraperitoneally and was euthanized 24 hours and 6 weeks after surgery.Another group received ADSCs intraperitoneally 6 weeks after surgery, and was euthanized at the 10 th week.About 2x10 5 cells were administered to each animal.

Evaluation of the renal function
Serum creatinine was measured by the modified Jaffé method.Serum urea was measured using the Labtest kit (Labtest, Lagoa Santa, MG, Brazil).

Morphology
The kidneys of the animals euthanized at the 6 th and 10 th weeks after ischemia were analyzed with the use of Masson and Picrosirius stains.For histological tests, the kidneys were fixed in buffered 10% formaldehyde for 24 hours, washed with 70% ethanol for 24 hours, and then immersed in paraffin.The histological slices were 4 µm thick.In order to assess the degree of expansion of the renal interstitium, the fraction of the renal cortex occupied by the interstitial tissue stained positively for components of the extracellular matrix (collagen) was evaluated quantitatively using Masson staining and a technique of counting consecutive points in microscopic fields, with a final magnification of 100x on a 176point grid.Picrosirius staining was measured with 20x magnification using the NIS-Element Nikon, program of microscopy elements, with at least 20 consecutive fields.Acute tubular necrosis (ATN) was evaluated by a blind reviewer in slices of kidneys stained with hematoxylin and eosin.

Bioplex
The assay kit Bioplex for use in mice with 23 plex of cytokines (Bio-Rad Laboratories, Inc., Hercules, CA, USA) was employed to test samples for the presence of 23 cytokines.The assay was read on the Bioplex matrix suspension system, and data were analyzed using Bio-Plex software Manager 4.0.Standard curves varied from 32.000 to 1.95 pg/mL.

Immunohistochemistry
Collagen I (diluted at 1:200; COL-1, Abcam, Cambridge, MA, USA), FSP-1 (diluted at 1:400, S100A4, DAKO), PCNA (diluted at 1:300; clone PC10, DAKO), and Hypoxyprobe (diluted at 1:500; 121 Turnpike Middlesex, Burlington, MA, USA) were carried out on histological slices fixed in paraffin.The slides were previously deparaffinized, rehydrated, and submitted to a solution for Tris-EDTA antigen recovery, with pH 9 and at 95°C.For the Hypoxyprobe, antigen recovery was performed with 10 mM citrate buffer, with pH 6, at 55°C.The activity of endogenous peroxidase was blocked by 3% hydrogen peroxide, besides the blockage with the Protein Block Solution (DAKO, Carpinteria, CA, USA).Next, the slides were incubated with a primary antibody or negative control reagent, followed by incubation with Envision (DAKO), using sequential 30-minute incubations at room temperature.Staining was completed by a 1 to 3-minute incubation with 3.3-diaminobenzidine (DAB) + chromogenic substrate, which resulted in a brown precipitate in the presence of antigen.Counter-staining with hematoxylin was performed.For the analysis of tissue hypoxia, intraperitoneal HypoxyprobeTM_1 (pimonidazole HCl) was administered, at the dose of 60 mg/kg of body weight, approximately 25 minutes before euthanization.After inoculation, HypoxyprobeTM_1 (121 Turnpike Middlesex, Burlington, MA, USA) was distributed throughout all tissues, but it adhered only to certain proteins from cells that had an oxygen concentration lower than 14 micromolar -the equivalent of a partial pressure pO2 = 10 mmHg at 37ºC.

Statistical analysis
Data were expressed with mean ± SD.The differences between the two groups were assessed as to statistical significance with Student's t-test and ANOVA.Differences were considered significant for p < 0.05.

ADSC isolation and characterization
ADSCs were characterized by immunophenotypical assays using a few cell surface markers described in literature (20) .CD34, CD73, CD105, and CD45 were evaluated.The expression of CD73 was 24.9 ± 0.9%; CD105: 71.2% ± 19.7; CD34: 6.0 ± 3.8%; and CD45: 6.9% ± 1.9.These results were expected in regard to the immunophenotypical profile described for ADSCs.First passage cells were characterized by differentiation into adipocytes in parallel assays.In cells differentiated into adipocytes, it was possible to visualize red drops of lipids inside the cells stained with Oil Red.The differentiation into osteocytes was performed, and after 28 days, calcium deposits were observed using Von Kossa staining (data not shown).

Treatment with ADSCs leads to tissue and systemic immunomodulation
In this model, serious ischemia and reperfusion occurred in only one kidney.However, due to the prolonged time of ischemia (1 hour), renal functional parameters were altered.Serum levels of urea were increased in animals subjected to severe ischemia when compared to Sham animals.Treatment with ADSCs displayed inferior renal dysfunction (Figure 1A).These results correlated with ATN.Kidneys submitted to ischemia after 24 hours showed a larger area of necrosis (Figure 1B), while ATN was decreased in animals treated with ADSCs.Additionally, the pattern of regeneration was superior in animals treated with ADSCs (Figures 1C and 1E) confirmed by PCNA marking (Figures 1F to 1H).
Inflammation plays a significant role in the results of renal ischemia and reperfusion.In addition, it is known that ADSCs lead to immunomodulation.In this sense, we analyzed the expression of mRNA of various cytokines after treatment with ADSCs.The expression of IL-6 mRNA was reduced in the renal tissue of animals treated with ADSCs (Figure 1I).The expression of TNF-α and IL-1β mRNA was also quantified, but no statistical differences were noted (Figures 1J and  1K), although the animals treated with ADSCs showed greater expression.Surprisingly, the expression of the mRNA of anti-inflammatory cytokines IL-4 and IL-10 was increased in the renal tissue of animals treated with ADSCs (Figures 1L and 1M).
In order to verify the presence of ADSCs within the kidneys after treatment, the expression of SRY mRNA was amplified by PCR in real-time (Figure 1U).No intensities of SRY mRNA were detected, indicating the absence of these cells within the tissue at this time of the study.

ADSC ceases the progression of fibrosis resulting from severe ischemia
After 6 weeks, in this model of reperfusion ischemia, it was noted that the kidney submitted to ischemia was smaller in comparison to the contralateral kidney in non-treated animals.Surprisingly, the ischemic kidney of the animals treated with ADSCs was not reduced in size (Figure 2A).Even so, the levels of creatinine and urea did not change under treatment with ADSCs, possibly due to the presence of the contralateral kidney, which gradually compensates the functional loss of the damaged kidney (Figures 2B and 2C).This reducedsize kidney also showed a larger area of fibrosis, as demonstrated with Masson trichrome and Picrosirius staining (Figures 2D to 2F).No relevant fibrosis was observed in the sham group.
We also quantified the expression of mRNA of a few profibrotic molecules in order to ratify our data as to collagen deposition.Type 1 collagen (Col-1), vimentin, and connective tissue growth factor (CTGF) were evaluated in renal tissue after 6 weeks.Once again, a reduced expression of the mRNA of these molecules was noted in the renal tissue of the animals treated with ADSCs (Figures 3A to 3C).In addition, in order to confirm these results, protein analyses were performed by means of immunohistochemistry for FSP-1 (fibroblastspecific protein 1) and Col-1 (Figure 4).

Immunomodulation of inflammation by ADSCs
As early as 24 hours after serious ischemia and reperfusion, a significant modulation of the inflammatory response was noted in animals treated with ADSCs.Since this reduced inflammation at the initial time point could be related to the progression of fibrosis, we investigated the inflammatory pattern inside the kidney and, systemically, 6 weeks after the treatment.Additionally, we quantified a few protective and angiogenic molecules related to tissue repair.In 6 weeks, the expressions in the kidney of proinflammatory cytokines, such as TNF-α and IL-6, were still reduced in the animals submitted to severe ischemia and treated with ADSCs (Figures 3D and 3E).Also, heme oxygenase 1 (HO-1) and the BMP-7/TGFb ratio were increased in the animals treated with ADSCs compared to those not treated (Figures 3F and 3G).Therapy with ADSCs also led to systemic immunomodulation.Serum levels of cytokines were reduced in animals treated with ADSCs.The levels of IL-1α, TNF-α, KC, and RANTES were significantly reduced in treated animals when compared to those not treated (Figures 3K, 3M, 3N, and 3O).IL-6 and other cytokines (IL-1β, IL-10, and IL-17) did not differ between the groups, although a tendency of reduction in animals treated with ADSCs was observed (Figure 3L and supplementary data).
Chronic hypoxia, due to rarefied capillaries, is a possible theory to explain the continued progression of renal fibrosis, since low oxygen pressure is an important factor for EMT.Therefore, we analyzed whether tissue hypoxia was reduced in renal tissue by using a probe that detects areas of low pO2.The ischemic kidneys of animals treated with ADSCs demonstrated areas of lower hypoxia (Figure 3J).Additionally, the expression of iNOS mRNA was increased in the kidneys of animals

Reversal of renal fibrosis after treatment with ADSCs
We tested the role of ADSCs in reversing an established fibrotic condition.ADSCs were administered 6 weeks after ischemia, and the animals were observed until the 10 th week.Amazingly, the kidneys of animals treated with ADSCs showed reduced areas of fibrosis, as per analysis with Picrosirius staining (Figures 5C  and 5D).A functional improvement was also noted and correlated with the reduction of these areas of fibrosis (Figures 5A and 5B).Confirming these data, the expression of vimentin mRNA was reduced in animals treated with ADSCs when compared to those not treated (Figure 6).The expression of Col-1 mRNA was also diminished in animals treated with ADSCs, but no significant difference was observed (Figure 6B).In addition, the immunohistochemical test for Col-1 and FSP-1 correlates with the studies of the genes analyzed (Figure 6E).The mRNA expression of protective cytokines such as IL-10 and BMP-7 increased after treatment with ADSCs (Figures 6C  and 6D).Once again, despite this functional and histological improvement, ADSCs were not found in the kidney, according to analysis of the expression of SRY mRNA (Figure 6F).

DISCUSSION
ADSCs were first described by Zuk et al. (27) in 2002.
Since then, studies with ADSCs have been growing exponentially.The regenerative properties of ADSCs are greatly attractive for many diseases (28) .Fibroproliferative diseases call our attention because the tissue destruction caused by deposition of extracellular matrix leads to functional loss of the organ.This is very important for the kidneys since the balance between cells and extracellular matrix determines their function.
CRD is emerging as a worldwide health problem.The plausible therapies for the final stage of renal failure today are dialysis and transplantation.Nevertheless, these treatments do not solve this problem (29) .Various comorbidities may cause CRD, but one aspect is always forgotten: persistent inflammation after ARF events (30,31) .Inflammation is the primary factor in the progression of renal fibrosis.It leads to the transformation of epithelial cells into myofibroblasts, known as EMT (4) .There is no consensus as to whether EMT occurs in vivo.However, the role of inflammation in scar formation is not questionable.Despite the functional improvement after acute kidney damage, a residual inflammation may contribute to epithelial stress that leads to fibrosis.In this model of serious unilateral ischemia, residual inflammation leads the kidney to decrease in size due to the cicatricial fibrosis formed.
The primary mechanism of action of adult stem cells is immunoregulation (23,24) .This occurs by means of a predominantly paracrine effect of the ADSCs, via a "touch and go" phenomenon in which these cells are attracted to the site of the lesion, secrete immune factors, and then leave that tissue or die.This may explain why ADSCs were not seen in the kidneys of treated animals.
In this project, we proposed that ADSCs may inhibit EMT by acting on inflammation.In 24 hours, the animals treated with ADSCs had better functional results correlated with immunomodulation, which is reflected after 6 weeks when smaller areas of fibrosis are observed.
In our results, over 10 weeks of analyses we were able to observe that, besides the fibrotic process, ADSCs were capable of reversing this condition of fibrosis.This opens an incredible opportunity for clinical treatment.We focus, then, on inflammatory results.Once again, the animals treated with ADSCs had low levels of inflammation, which might be reflected in the reduced progression of extracellular matrix deposition.
In summary, we propose that treatment with ADSCs may lead to a halt in the progression or even to the reversal of the fibrotic process by downregulation of the inflammation, via systemic administration.

Figure 1 .
Figure 1.Effects of administration of adipose tissue-derived stem cells after 24 hours of unilateral ischemia.
, despite significant differences not having been found in the expression of VEGF (Figures3H and 3I).

Figure 2 .Figure 3 .Figure 4 .
Figure 2. Effects of administration of adipose tissue-derived stem cells in mice after 6 weeks of unilateral ischemia.

Figure 5 .Figure 6 .
Figure 5. Effects of administration of adipose tissue-derived stem cells in week 6 after ARF, analyzed after 10 weeks.