Mesenchymal Stromal Cells Prevent Renal Fibrosis in a Rat Model of Unilateral Ureteral Obstruction by Suppressing the Renin-Angiotensin System via HuR

We studied Mesenchymal Stromal Cells (MSC) effects in experimental Unilateral Ureteral Obstruction (UUO), a fibrogenic renal disease. Rats were divided in 5 groups: sham, UUO, MSC treated-UUO, ACEi treated-UUO, MSC+ACEi treated- UUO. Data were collected at 1, 7, 21 days. UUO induced monocyte renal infiltration, tubular cell apoptosis, tubular atrophy, interstitial fibrosis and overexpression of TGFβ, Renin mRNA (RENmRNA), increase of Renin, Angiotensin II (AII) and aldosterone serum levels. Both lisinopril (ACEi) and MSC treatment prevented monocyte infiltration, reduced tubular cell apoptosis, renal fibrosis and TGFβ expression. Combined therapy provided a further suppression of monocyte infiltration and tubular injury. Lisinopril alone caused a rebound activation of Renin-Angiotensin System (RAS), while MSC suppressed RENmRNA and Renin synthesis and induced a decrease of AII and aldosterone serum levels. Furthermore, in in-vitro and in-vivo experiments, MSC inhibit Human antigen R (HuR) trascription, an enhancer of RENmRNA stability by IL10 release. In conclusion, we demonstrate that in UUO MSC prevent fibrosis, by decreasing HuR-dependent RENmRNA stability. Our findings give a clue to understand the molecular mechanism through which MSC may prevent fibrosis in a wide and heterogeneous number of diseases that share RAS activation as common upstream pathogenic mechanism.


Introduction
Renal fibrosis represents the common ending of several renal diseases. Experimental Unilateral Ureteral Obstruction (UUO) has provided extensive information on renal fibrogenesis. [1][2][3] Different studies have shown that UUO causes a sustained glomerular ischemia, that is associated with the activation of the Renin Angiotensin System (RAS). [4][5][6] Angiotensin II (AII) stimulates the production of nuclear NF-kB, which recruits macrophages that release reactive oxygen species [7,8] and Tumor Necrosis Factor α (TNFα), an inducer of apoptosis and tubular atrophy. [1,9] Moreover, inflammatory cells release Transforming Growth Factor β (TGFβ) that promotes epithelial mesenchymal transition of tubular epithelial cells, while fibroblasts synthesize stress fibres and deposit extracellular matrix causing progressive interstitial fibrosis. [10][11][12][13] Furthermore, recent studies give evidence that Renin binding with its (pro)renin receptor directly prompts the extracellular-signal-regulated kinase (ERK)-dependent production of TGFβ. Therefore, Renin has a fibrogenic effect that is independent of its enzymatic action on angiotensinogen. [14][15][16] Human antigen R (HuR) has been recently described to stabilize RENmRNA sustaining Renin production. In the same time it has been demonstrated that HuR expression was inhibited by IL10. [17][18][19] Currently, RAS is the only fibrogenic system targeted by drug therapy. In fact angiotensin-converting enzyme inhibitors (ACEi) and AII receptor blockers are used in clinic to slow down the progression of renal diseases. [20][21][22][23][24][25][26][27] MSC are isolated from several human tissues, can differentiate in multiple adult cell lineages, and display immunomodulatory properties. [28][29][30] When injected in animal disease models MSC migrate to sites of inflammation and injury and secrete paracrine mediators for tissue repair. [31][32][33][34][35][36] These pleiotropic activities have prompted the use of MSC for treatment of diseases where cell damage is accompanied by local and systemic inflammatory response [37][38][39][40] and for prevention of fibrosis in animal models and clinical studies. [41][42][43][44] Here we show that MSC prevent fibrogenic kidney injury in experimental UUO and compare their effects with those of lisinopril, an ACEi. Furthermore, we report on the potential therapeutic advantage of their combined use. Finally we show that MSC suppress RAS, inhibiting HuR by IL10 release.

Animals
61 male Wild-type Sprague Dawley (SD) rats (Charles River, Lecco, Italy) weighed 250-300 g, were used for animal model. While 6 transgenic Enhaced Green Fluorescent Protein (EGFP) SD rats (Japan Slc, Hamamatsu, Japan) were used to expand MSC. [45] Rats were housed at constant temperature (20 C°) and humidity (75%) under controlled light cycle, with free access to water and standard chow diet.

Isolation and Colture of MSC
Transgenic EGFP SD rats were sacrificed by CO 2 inhalation. Marrow cells were flushed from femurs and tibias and mononuclear cells were isolated by density gradient centrifugation (Ficoll 1077 g/ml; Lympholyte, Cedarlane Laboratories Ltd.) counted and plated in 75 or 175 cm 2 tissue culture flasks at a density of 160000/cm² in αMEM supplemented with 10% murin MesenCult (Voden, Milan, Italy) and 1% antibiotic-antimycotic (Sigma-Aldrich, Saint Louis, US). Cultures were maintained at 37°C, 5% CO 2 in a humidified atmosphere. After 48 hours, non-adherent cells were removed and culture medium was replaced twice a week. After reaching 80% confluence, MSC were harvested using Trypsin-EDTA (Lonza, Copenhagen, Denmark), and propagated at 4000 cells/cm².

In vivo experiments
In vivo experiments were performed according to guidelines of our ethical committee for animal studies and approved by ministerial ordinance n°36262. SD rats were anesthetized using intramuscular injection of Zolazepam-Tiletamine (30 mg/ Kg) (Zoletil). After the induction of anesthesia a flank incision was made and the right ureter was ligated with 5/0 silk suture to induce obstruction. Animals were divided in 5 groups: Sham: 8 rats sham operated. Vehicle: 8 rats received 1 ml of saline solution after UUO at day 0 via tail vein. MSC: 8 rats received one dose of MSC (3X10 6 ) after UUO at day 0 via tail vein. ACEi: 8 rats received lisinopril in the drinking water (100 mg/L) from days 1 to 21. ACEi+MSC: 8 rats received one dose of MSC (3X10 6 ) on day 0, and ACEi from days 1 to 21. Rats were sacrificed at day 7 and 21 after ureteral ligation. Blood samples were collected at 7 and 21 days, centrifuged and maintained at -20°C until analyses were performed.
Another set of experiments was performed using 3 groups of rats sacrificed 1 day after ureteral ligation: Sham: n = 5, Vehicle: n = 8, MSC: n = 8. Soon after sacrifice kidneys were removed, half was fixed in 10% neutral-buffered formalin, half frozen in liquid nitrogen.

Biochemical measurements
Serum creatinine levels were measured by an automated method (Abbott Laboratories, Chicago, Illinois, US). Serum AII levels were quantified by ELISA (LifeSpan BioSciences, Newton, US). Serum aldosterone levels were measured by chemiluminiscent assay (Liaison Aldosterone assay, DiaSorin, Vercelli, Italy).

RNA Extraction and Real time PCR
Total RNA was extracted from renal tissue and cultured cells (HK2 cell line and macrophages) using respectively trizol method and guanidine based RNeasy 1 Mini Kit (QIAGEN Venlo, Holland). RNA was treated with DNase from RNase-Free DNase Set (QIAGEN) and dissolved in nuclease free water. Extracted RNA was tested for quantity and integrity by spectrophotometric analysis (NanoDrop, Thermo Scientific, Watham, US). A total of 1 μg of RNA per condition was reverse transcribed into cDNA through 1st Strand cDNA Synthesis Kit for RT-PCR (AMV) (Roche Applied Science, Basel, Germany). cDNA was used to perform real-time PCR analysis in 96-well optical reaction plates, using ABI prism 5700 (Applied Biosystems, Watham, US) and the 5-exonuclease assay (TaqMan technology) in a volume of 25 μl reaction containing Taqman Universal Master Mix, optimized concentrations of FAM-labeled probe, and specific forward and reverse primers for βactin, Renin and HuR (Applied Biosystems, Watham, US; catalogue number Rn00667869_m1, Rn00561847_m1 and Rn01403240_m1, respectively). Water replacing cDNA was included in real-time PCR as controls. The results were analyzed using a comparative method, and values normalized to the βactin expression and converted into fold change.

Renal Histopathology
2.8.1 Apoptotic nuclei. Renal sections were dewaxed in xylol, passed in a decreasing series of alcohol, and finally rehydrated with distilled water. For permeabilization the tissues sections were exposed to distilled water containing 0.1%Triton X-100, 0.1% citrate sodium, then incubated with proteinase K in Tris/HCl10mM, pH 7.4-7.8. After washing in PBS, sections were incubated with 20 ml of TUNEL reaction mixture (In Situ Cell Death Detection Kit, Fluorescein, Roche) for 1 hour at 37°C, in a dark humid chamber. After washing in PBS, specimens were mounted in Vectashield Mounting Medium with DAPI (Vector Laboratories, Burlingame, US) and examined with fluorescence microscope Nikon E400 with CCD camera and software imaging analysis Cell-R. In order to assess apoptotic nuclei we examined 10 non-overlapping fields per renal section (X400) per animal and expressed them as means ± SD.
2.8.2 Renal Morphology. 3 μm thick sections of formalin-fixed kidney were examined by two investigators in double blind fashion, using an Olympus IX8 microscope connected with a CCD camera and software imaging analysis Cell-R.
2.8.3 EGFP, ED1 Ag and Fibroblast Specific Protein 1 (FSP1) expression. EGFP, ED1 Ag and FSP1 renal expression were studied by immunohistochemistry (IHC) in formalin fixed tissue. Three-μm thick sections of paraffin embedded tissue were collected on poly-L-lysinecoated slides (Dako, Glostrup, Denmark), they were dewaxed in xylol, passed in a decreasing series of alcohol, and finally rehydrated with distilled water. Endogenous peroxidase was blocked with H 2 O 2 /methanol 3.7% vol/vol for 10 minutes followed by H 2 O. After 3 washings in PBS sections underwent microwave antigen retrieval, then were exposed overnight at 4°C to the following antibodies: 1) monoclonal mouse anti-green fluorescent protein antibody IgG1 (Chemicon International, Billerica, US), 2) monoclonal mouse anti-rat ED1 (Serotec Ltd, Kidlington, UK), 3) polyclonal anti S100-A4 (Dako). After 3 washings in PBS the immunocomplex was visualized with the biotin-streptavidin-peroxidase complex and 3,3-diaminobenzidine (Dako). Sections were faintly counterstained with Harris hematoxylin. Negative controls included both omission of the primary Ab and substitution of IgG for primary antibodies. The positive controls for EGFP were kidney sections of SD-EGFP. In 10 renal sections per kidney we counted EGFP, ED1 and FSP1 positive cells/HPF. Since EGFP cells were spontaneously fluorescent we studied EGFP renal expression also by fluorescence method. Renal sections were air dried and post-fixed in ethanol 70%, 95%, 100% (3 min). After washing in PBS, specimens were mounted in Mowiol Mounting Medium (Sigma-Aldrich) and examined with fluorescence microscope Nikon E400 with CCD camera and software imaging analysis Cell-R.

Statistical analysis
ANOVA followed by the Newman-Keuls test were used to compare not parametric variables and Student t test for the means.

Rat MSC characterization
EGFP-MSC showed the typical spindle shape morphology (Fig 1, panel 1) and resulted positive for CD49, CD90, and CD29, while they were negative for CD45 and CD11b. (Fig 1, panel 2) MSC exhibited the capacity to differentiate in vitro to osteoblasts and adipocytes as confirmed by histological staining. Microscopic examination of stained cells demonstrated the presence of mineralization nodules and fat droplets. (Fig 1, panel 3)

In vivo experiments
Through the study period, no significant differences were observed in body weight, water and food intake among the 5 groups of rats. No side effects were documented.

MSC localization
In EGFP-MSC treated rats, few cells were tracked (0-5 per section/HPF), on kidney sections obtained at day 1, and they were located in glomeruli, tubules and interstitium. (Fig 1, panel 4)

Renal function
At 7 and 21 days serum creatinine levels were significantly higher in UUO compared to sham (p<0.05). There was no difference among groups of obstructed rats, nor between day 7 and day 21 (Fig 2) 3.5 RENmRNA expression, Renin tissue levels, AII and aldosterone serum levels After 7 days of UUO, RENmRNA expression significantly increased in renal tissue of all groups except for MSC D7 (p<0.001), while in MSC and ACEi+MSC groups RENmRNA expression decreased in comparison to vehicle and ACEi groups (p<0.05), suggesting a MSC inhibitory effect on RENmRNA expression. (Fig 3, panel A) Accordingly, at day 7 Renin tissue levels increased in vehicle and ACEi rats compared to sham, while in MSC and ACEi+MSC groups they were significantly decreased compared to vehicle and ACEi rats (p<0.05). (Fig 3, panel B) As sign of RAS activation, after 7 days AII and aldosterone serum levels significantly increased in UUO rats (p<0.005). MSC, ACEi and combined therapy induced a significant decrease in AII and aldosterone serum levels (p<0.005). (Fig 3 panels C and D, respectively)

Monocyte infiltrate
As shown in Fig 4 panel B, UUO induced a significant MC infiltration increase in renal tissue (p<0.0001). After 7 days of ligation, MC were significantly reduced by MSC and ACEi+MSC (p<0.001). Moreover combined therapy significantly reduced MC infiltration compared to ACEi (p<0.05). At 21 day, MC infiltration was significantly reduced by all treatments (p<0.0001). (Fig 4, panel B)

Apoptosis
Tunel tubular positive cells increased significantly after 7 and 21 days of ureteral ligation in vehicle rats compared to sham operated rats (p<0.005). At day 7, ACEi+MSC treatment significantly decreased tubular cell apoptosis (p<0.05). At day 21 MSC and ACEi+MSC treatment significantly decreased tubular cell apoptosis (p<0.005). (Fig 4, panel D)

Tubular damage
Tubular damage was evaluated at day 7 by measuring tubular diameters and at day 21 by enumerating damaged Tubular Basement Membrane (TBM). Tubular diameter was significantly increased in UUO compared to sham (p<0.0001). Renal sections of MSC, ACEi and ACEi +MSC treated rats showed tubules with significantly smaller diameters compared to vehicle (p<0.05). (Fig 5, panel B) The number of tubules with damaged TBM was significantly reduced by combined therapy compared to vehicle and MSC alone (p<0.01). (Fig 5, panel C)

Renal tissue TGFβlevels
After 7 days of UUO, tissue TGFβ levels increased in vehicle group. In all treated groups, TGFβ levels were significantly lower (p<0.05). (Fig 6, panel B)

Renal fibrosis
Fibrosis was defined both as FSP1 positive cells/HPF and Masson's trichrome positive area/ HPF (X400). After 21 days of obstruction FSP1 positive cells increased significantly in non treated animals (p<0.001). (Fig 6, panel C) A significant decrease of fibroblast amount was observed following ACEi, MSC and combination therapies (p<0.005). Masson's trichrome staining showed that fibrosis score was significantly (p<0.001) less severe in rats treated both with MSC and ACEi+MSC compared to untreated rats at day 21. (Fig 6, panel D) 3.11 Renal Human antigen R (HuR) mRNA, RENmRNA expression and Renin levels at day 1 HuRmRNA, RENmRNA expression and Renin levels increased significantly 1 day after ureteral ligation in non treated animals (p<0.001). MSC treatment inhibited HuRmRNA, RENmRNA expression and Renin levels. (Fig 7)

In vitro experiments
3.12.1 MSC/HK2 co-coltures. RENmRNA expression, as expected, was significantly higher in HK2 cultured in high glucose (HG) medium than in low glucose (LG) medium (p<0.05). (Fig 8, panel A) Presence of MSC in HK2 HG cultures reduced RENmRNA expression in a dose dependent manner. At 1:2 MSC/HK2 ratio RENmRNA expression was similar to HK2 LG expression. (Fig 8, panel A) 3  MSC in co-culture reduced RENmRNA expression in Mϕ, significantly at 1:2 ratio (p<0.005). (Fig 8, panel B) 3.12.3 MSC modulate RENmRNA expression via IL10. In order to investigate the mechanism through which MSC suppress RENmRNA expression, we performed in vitro culture experiments in presence of IL10 and anti-IL10 blocking antibody. Addition of IL10 to HK2 HG cultures significantly decreased RENmRNA expression (p<0.001). (Fig 8, panel C) While addition of anti-IL10 blocking antibody in MSC/HK2 HG co-culture reversed MSC suppression on RENmRNA expression. (Fig 8, panel D) Similarly RENmRNA expression in Mϕ cultures additioned with IL10 was significantly lower than in Mϕ without IL10 (p<0.05). (Fig 8,  panel E) The addition of anti IL10 blocking antibody in co-cultures MSC/Mϕ reversed MSC suppressive effect. (Fig 8, panel F)

Discussion
It is well documented that acute UUO is followed by a prompt renal vasodilatation that turns into a persistent vasoconstriction whithin 24 hours. [4][5][6] The ensuing ischemia is associated with the progressive development of tubular atrophy and interstitial fibrosis that are a   proved effect of increased AII release resulting from enhanced Renin production. [2,3] In addition, recent studies have shown that Renin has a fibrogenic effect that is independent of its enzymatic release of AII. In fact, the interaction of Renin with its receptor activates the production of TGFβ transduced by the ERK system. [14][15][16] MSC have been shown to have an antiiflammatory and antifibrogenic effect in several models of injury to various tissues, including the kidney. [35][36][37]46] In the present study we have investigated whether MSC prevent UUO induced fibrosis and we have tested whether MSC operate this effect through Renin. Furthermore, we have compared the results obtained following MSC infusion with those of ACEi treatment and the combination of both in order to add therapeutic advantage. Our results show that MSC decrease tubular cell apoptosis and infiltrating monocyte number and preserve tubular integrity preventing luminal dilatation and disruption of the basement membrane. These effects were associated with a significant decrease of fibrotic renal tissue, that did not result in a significant effect on renal function, as indicated by serum creatinine levels. These findings may be argued as contradictory, but a sound explanation is provided by simply considering that the loss of renal function associated with fibrosis in the obstructed kidney is compensated by increased GFR of the contralateral organ. [47] Of course, prevention of fibrosis as operated by MSC or ACEi treatment limits the loss of function in the obstructed kidney, but does not induce any compensatory increase in controlateral function.
Kidney protection afforded by MSC was comparable to that provided by ACEi, while a further decrease of renal inflammation and interstitial fibrosis was observed with the combined use of MSC and ACEi. Our results provide more informations on the MSC therapeutic potential described in previous studies, where it has been shown that MSC injected into obstructed kidney reduce the expression of collagen synthesis, fibroblast activation [42] and infiltrating macrophage number. [37][38][39][40][41]43,48,49] However the main innovation of our study was to shed some light on the possible mechanism(s) that account for MSC effects. It is well known that in acute and chronic models, MSC may act by paracrine action rather than a transdifferentiation into renal resident cells. This hypothesis is supported by our finding that very few EGFP-MSC were tracked in the kidney.
Moreover, we observed that after UUO, MSC inhibited expression of RENmRNA, Renin, AII and aldosterone, suggesting that MSC may antagonize RAS. It has been shown that RAS activation induces recruitment of inflammatory cells in several disease models including UUO [7,8], with release of inflammatory, pro-apoptotic and fibrogenic cytokines that enhance tissue damage. [10,[50][51][52][53] Therefore, RAS inhibition observed in MSC treated rats may also explain prevention of macrophage infiltration.
We also found that MSC abated TGFβ that plays a crucial role in generating fibrosis, because it stimulates matrix proteins synthesis, decreases matrix degradation, stimulates fibroblast proliferation and synthesis of stress fibres, promotes epithelial-mesenchymal transition of tubular cells. [11][12][13] This finding may be a consequence of Renin inhibition by MSC. A question raised by our study is how MSC affect Renin release. In renal tissue, granular cells of afferent arteriole are the main source of Renin, but RENmRNA is expressed in tubular cells [54,55] and infiltrating macrophages [56], too.
Since MSC prevent macrophage infiltration [44,48,49], it is reasonable to assume that one mechanism by which MSC inhibited RAS was by blocking the source of Renin represented by macrophages. [57] However, the behaviour of RENmRNA expression in rats treated with ACEi and MSC suggests that MSC directly downregulated Renin production by renal cells. We have shown that the increase in RENmRNA and Renin that occurs in ACEi treated rats is suppressed by MSC. This suppression takes place while the inhibition of AII caused by ACEi persists, so that MSC suppress Renin by a mechanism that does not involve the AII-evoked feedback. The evidence that MSC and ACEi operate RAS inhibition in a different way provides a rationale for using them in combined therapy. Actually in our study their combination strength their anti-fibrogenic effect.
In order to understand MSC mechanism, we focused on: a) Renin release, first step of RAS activation, b) HuR expression, that sustains Renin production, stabilizing RENmRNA [17][18][19]58,59] c) IL10, cytokine known to be produced by MSC [54][55][56]60,61] and able to inhibit HuR expression. [17][18][19] For the first time we proved in ex vivo experiments that MSC reduce REN and HuR mRNA expression in obstructed kidney, already one day after UUO. To understand whether MSC inhibited REN and HuR mRNA expression via IL10, we performed in vitro experiments and demonstrated that MSC downregulated RENmRNA in tubular cells [62,63] and in macrophages [57]. IL10 inhibited RENmRNA expression in activated tubular cells and macrophages and the addition of anti IL10 blocking antibody in co-cultures of MSC/tubular cells and MSC/ macrophages reversed MSC effects. We also measured IL10 in MSC culture surnatants but we found only slightly detectable levels (data not shown), perhaps due to the fast use of IL10 by the cells.
In conclusion, we have shown that MSC injection in UUO prevents inflammation, tubular cell apoptosis and fibrosis. These effects are associated with the suppression of RAS, macrophage infiltration and TGFβ expression. These findings allow to reconstruct a rational sequence of events that may represent a RAS inhibitory pathway driven by MSC. We confirm that RAS inhibition is associated with HuR suppression, and give evidence that MSC suppress HuR and thus inhibit RAS. IL10 released by MSC may be responsible for HuR suppression.
The present study is a further demonstration that MSC mitigate injury and/or favour repair by delivering soluble factors that in a paracrine or even endocrine manner modify the cell behaviour favouring kidney recovery.